In the banking sector, this research projectwill identify factors that affect the acceptance of chatbotconsulting services by customers living and working inVietnam.. Research results ha
Introduction
l.L The urgency of the subject
In Vietnam, digitalization is increasingly transforming various sectors, including commerce, finance, banking, education, and healthcare The government is actively fostering digital economic growth, particularly in the banking and finance sectors, through strategic policy measures, such as Decision No 986/QD-TTg issued in August.
In August 2018, the Prime Minister approved the Vietnam Banking Industry Development Strategy to 2025, alongside the Banking Industry Digital Transformation Plan, aimed at enhancing digital banking models and improving customer experience This initiative, outlined in Decision 810/QD-NHNN by the State Bank, focuses on automating processes and optimizing operations through advanced technology It sets the stage for the government to enhance legal frameworks and invest in technological infrastructure, thereby fostering the development of digital technology within the banking sector After two years of implementation, significant achievements in digital transformation have underscored the critical role of AI in advancing the finance and banking industry.
Many banks and financial institutions are rapidly investing in digital experiences for customers, recognizing it as a key trend and competitive advantage in the financial sector There is a strong emphasis on customer care and consulting services, particularly through the enhancement of Chatbots that leverage advanced AI technologies to optimize operations management in banking.
Table 1: Current status of chatbot applications in Vietnamese banks
Bank Chatbot Date of application
Source: Compiled by the author during the research process
14 Viet Capital Bank Zalo OA 9/2021
This research offers insights into consumer interactions with chatbots, particularly within the context of Vietnam's emerging banking industry While chatbots represent a significant advancement in customer care and online payment solutions, their application remains suboptimal due to a lack of diverse retail banking products tailored for individual customers—who are the primary audience in need of chatbot assistance Survey results indicate that 68% of users plan to continue utilizing chatbots, highlighting their potential in the banking sector However, Vietnamese banks have yet to prioritize the development of this technology, especially in light of the recent advancements in artificial intelligence, such as the introduction of Chat-GPT in late 2022, which has intensified competition among global tech companies Despite the advantages chatbots offer, there is a notable gap in research regarding the factors influencing customers' willingness to use virtual assistants in Vietnam's commercial banking landscape.
RO1: Identify factors affecting customers' ability to accept Chatbots in using consulting services in Vietnamese banks.
RO2: Assess the impact of factors on Chatbot adoption.
The introduction of AI chatbots across various communication platforms, including mobile apps, websites, and social media pages, has transformed customer support These intelligent tools assist users with essential tasks such as consulting services, updating information on interest rates and exchange rates, checking account balances, and applying for credit cards.
The advancement of AI technology has empowered chatbots to utilize natural language processing (NLP), facilitating more complex interactions with users and expanding their applications in e-commerce, including financial consulting and customer service (Heo and Lee, 2018) Moreover, these chatbots benefit from continuous learning and round-the-clock availability, allowing them to enhance their intelligence through customer interactions, thanks to AI's self-learning capabilities.
As technology becomes more and more developed and optimized, will customer trends in choosing the type of consulting change? Hence, the research on the topic
The article "Characteristics of Chatbots Affect Customers' Decisions to Use AI Consulting Services in the Banking Sector: The Case of Vietnam" examines the factors influencing customers' transitions from human to Chatbot virtual consultants in Vietnam's banking industry It synthesizes feedback from frequent bank service users to analyze their preferences Based on this analysis, the team proposes strategies for effectively developing banking Chatbots in Vietnam to enhance customer attraction and satisfaction.
RO3: Propose recommendations to relevant parties (government, credit institutions,
Research Methods
Object and scope of research
Research object: factors affecting the ability of customers to accept the use of consulting services using AI technology in the banking sector.
Research scopes: To achieve the above research goal, the survey was conducted in
In September 2023, a survey was conducted involving 477 participants residing and working in Vietnam, primarily focusing on individuals engaged in the economic sector The respondents included a significant number of people who have utilized chatbot consulting services in the banking industry.
Literature review and hypothesis development
2.1 Literature review and research gap
In their research, "Chatbots or Me? Consumers' Switching Between Human Agents and Conversational Agents" (Li and Yang, 2023), the authors conducted a survey examining how consumer demographics, such as gender, age, education, and prior chatbot experience, influence preferences for human agents versus chatbots They also assessed various chatbot capabilities, including empathy, adaptability, and personalization, based on customer reviews However, the study faced limitations due to common method bias and analytical method bias, as identified by Harman (1967), stemming from the reliance on self-report measures and the use of SPSS 26 software for data analysis.
In their 2023 study, Hsu and Lin explored user satisfaction and loyalty concerning customer service chatbots by examining personal profile variables such as age, education, and previous chatbot experiences, alongside customer satisfaction factors like efficiency, completion, and security Their findings revealed that both response quality and conversational quality of AI chatbots significantly influence user satisfaction However, the applicability of these results is limited to Taiwan, a nation with a mobile penetration rate of 120% (Kemp, 2022) and a culture that readily embraces innovative technologies Furthermore, the research focused on general chatbot services without delving into specific examples from various fields.
The research article “Can Chatbots Satisfy Me? A mixed-method comparative study of satisfaction with task-oriented chatbots in mainland China and Hong Kong” (Liu et al., 2023) explores customer satisfaction and usage intentions regarding chatbots in these regions Utilizing a framework based on the Delone and McLean information systems model alongside privacy concerns, the study reveals that factors such as suitability, completeness, pleasure, and assurance significantly influence usage intention in both areas Notably, only the Chinese region identified response time and empathy as additional factors affecting satisfaction Interestingly, privacy concerns did not impact satisfaction in either region The study primarily focuses on general chatbot research and collects data from a specific demographic group.
In their 2023 research article, "The Effects of Chatbot Characteristics and Customer Experience on Satisfaction and Continuance Intention Toward Banking Chatbots: Data from Vietnam," Xuan Cu Le and Tran Hung Nguyen conducted a web-based survey targeting Chatbot users in the Vietnamese market The study highlights how AI consulting service characteristics impact customer experience and behavioral outcomes, such as satisfaction and the intention to continue using banking chatbots By consulting banking industry experts and surveying a small sample group, the authors validated the relevance of their data The positive findings indicate that the research is significant and can serve as a valuable reference for future studies.
In their research paper titled "Investigating the Factors of Customer Experiences Using Real-Life Text-Based Banking Chatbot: A Qualitative Study in Norway," Petersson, Pawar, and Fagerström introduced an innovative method for gathering data, complementing traditional surveys with qualitative insights.
In a 2023 study, eight participants engaged with banking Chatbots to perform tasks ranging from simple credit card registration to complex loan applications Follow-up interviews revealed that the human-like features of Chatbots and their ability to understand questions significantly enhance customer experience However, miscommunication issues negatively affect interactions, even with straightforward tasks The findings suggest that banks must educate customers about the limitations of AI capabilities and the security measures in place for complex transactions To successfully implement and improve Chatbot services, banks should prioritize consumer experience in their development strategies.
The theory of consumer behavior is a key concept in economics that explains how individuals allocate their limited resources among various goods and services to maximize their satisfaction within budget constraints It assumes that consumers have rational preferences and make consistent decisions based on these preferences and constraints Central to this theory are indifference curves, which illustrate combinations of goods that provide the same level of satisfaction, alongside budget constraints that reflect consumers' income and prices By examining the relationship between indifference curves and budget constraints, this theory predicts how changes in price and income influence consumers' purchasing decisions.
Technology Acceptance Model (TAM) is a theoretical information system in the form of a mode] that guides users in using technology and accepting its use Fred David
The Technology Acceptance Model (TAM), introduced in 1989 and based on the Theory of Reasoned Action (TRA), posits that the acceptance of an information system hinges on two key factors: perceived usefulness and perceived ease of use When users recognize technology as beneficial and user-friendly, they are more likely to embrace it Additionally, TAM highlights that external influences, such as peer support, individual psychology, and prior experience, can significantly shape users' behavior towards technology adoption.
TAM2 research model (Venkatesh and Davis, 2000):
The Theory of Reasoned Action (TRA), created by psychologists Ajzen and Fishbein, predicts behaviors based on attitudes but lacks specific measures for technology use To address this gap, Davis developed the Technology Acceptance Model (TAM), which builds on TRA to better assess technology acceptance Despite its advancements, TAM had limitations in theoretical models and measurement scales Consequently, TAM2 was introduced as an extension to enhance predictive power and explore the relationship between technology adoption and organizational productivity, while also considering additional factors influencing user perceptions of technology.
This study examines the factors influencing customers' perceived usefulness and ease of use of chatbots, incorporating variables such as information quality, service quality, system quality, and privacy and security By extending the Technology Acceptance Model (TAM), the research demonstrates that information quality, service quality, and system quality significantly impact users' perceptions of chatbots' helpfulness Additionally, it highlights that privacy and security considerations are crucial for users when opting for chatbots over traditional consultants in the banking sector.
This research investigates how a lack of technological knowledge influences customers' willingness to use chatbots for banking transactions By incorporating the variable of technology literacy, the study aims to understand how it affects customers' perceptions of chatbots as useful and user-friendly, ultimately impacting their decision to adopt this technology.
2.4.1 Perceived ease of use - PE
Perceived ease of use (PE) is a key indicator of user preference, highlighting that technology should be easily accessible without the need for advanced skills or significant effort from the user (Baber, 2019; Davis, 1989) According to the Technology Acceptance Model (TAM) and the Unified Theory of Acceptance and Use of Technology (UTAUT), users are more likely to embrace new technologies that provide a user-friendly experience (TAM; Davis, 1989) Therefore, we propose the following hypothesis:
*Hla: Perceived ease of use (PE) positively impacts Intent to use (IN) chatbot technology of customers in the hanking sector.
According to technology acceptance theory, users are more likely to perceive a new technology as helpful when they find it easy to use This observation leads to the hypothesis that ease of use significantly influences users' acceptance and perceived usefulness of technology.
*Hlb: Perceived ease of use (PE) has a positive impact on customers’ Perceived usefulness of Chatbot technology (PU).
Perceived usefulness of technology increases when that technology brings significant benefits and work efficiency to users (Min, s., So, K K F., & Jeong, M.,
In 2019, it was established that systems and technologies designed to save users time and fulfill their needs significantly enhance the likelihood of adoption In the banking sector, chatbot technology enables users to efficiently manage their transactions and credit activities, reducing the necessity for in-person visits to the bank This leads us to hypothesize that the integration of chatbots will positively impact user engagement and satisfaction.
[H2: Perceived usefulness (PU) positively affects Intent to use (IN) chatbot technology in the hanking sector.
Chatbots deliver essential information tailored to user needs, making the quality of this information crucial for their effectiveness and user-friendliness When customers receive valuable and relevant information, their perception of the chatbot's usefulness and ease of use significantly improves.
* H3a: Information quality (IQ) positively influences users' Perceived usefulness (PC).
* H3b: Information quality (IQ) positively influences users' Perceived ease of use (PE).
Research Methods
The research team employs established scientific methods by utilizing the TAM2 model (Venkatesh and Davis, 2000) as a theoretical foundation Additionally, they incorporate key variables such as Information Quality and System Quality, drawing insights from the study conducted by Liu et al (2023).
The quantitative research method involves creating a foundational model and gathering data from research participants through a Google Form questionnaire This data is then coded, cleaned, and assessed for scale reliability, followed by testing the model and hypotheses The analysis is conducted using IBM SPSS Statistics 26 and SmartPLS 3 software for thorough quantitative evaluation.
Basic statistical method: the measurement scale of the variables is inherited from previous research articles and the topic uses a 5-level Likert scale (1 - strongly disagree,
2 - disagree, 3 - neutral, 4 - agree, 5 - completely agree).
Check the reliability of the scale by analyzing Cronbach's Alpha coefficient.
The questionnaire will be structured according to the model suggested by the author group, comprising two sections: demographic inquiries and factors related to the model These factors include perceived usefulness (PU), perceived ease of use (PE), information quality (IQ), service quality (SeQ), system quality (SyQ), privacy and security (PS), intention to use (IN), and technology literacy (LT).
Questionnaire on factors related to the model
IQ1 Banking chatbot information is provided fully upon request
1Q2 Information provided regarding the problem
IQ3 Recommendation information tailored to personal data
ScQl Chatbot provides timely information
SeQ3 Chatbot gives natural feedback like a human
SyQI Chatbot updates all new trends
SyQ2 Chatbot has an easy-to-use, accessible interface
SyQ3 Chatbot helps me customize my financial experience to my preferences
PU1 Using a chatbot would be helpful for me
PU2 Using chatbot will be more convenient for me
PU3 Using Chatbot helps me find information faster
PU4 Using Chatbot makes me spend less time searching for information
Perceived ease of use (PEU)
PEU1 Using Chatbot is an easy way to find information
PEU2 Easy to understand how to use Chatbot
PEU3 Using Chatbot will be very easy
PEU4 Learning how to use Chatbot is very easy for me
PS I Information is highly confidential
PS2 Privacy and security policies are transparent and public
PS3 I feel the chatbot is trustworthy
LT1 1 have experienced AI applications
LT2 I have basic knowledge about technology
LT3 I can grasp important digital technologies
Intention to use chatbots (IN)
INI I will recommend using chatbots to my friends
IN2 I intend to use chatbots regularly
1N3 1 will use chatbot when there is a problem
A survey was conducted to gather data from users of banking AI chatbot services in Vietnam using Google Forms The questionnaire was divided into two sections to enhance the reliability and quality of the information collected The first section focused on the respondents' personal information, such as gender and industry group, while the second section aimed to identify factors influencing customers' perceived value of chatbot usefulness in the Vietnamese banking sector.
Table 2: Characteristics of the survey sample
From 5 million VND to under 10 million VND 100 28.6
From 10 million VND to under 20 million VND 84 24.0
6 Frequency of using Chatbot in the last 1 month
Source: Compiled by the author during the research process
Results
A survey conducted with 477 participants using Google Forms revealed that 127 respondents, or 26.62%, had never utilized AI chatbots in banking consulting services, leading to the exclusion of their responses Consequently, 350 valid samples remained, representing approximately 73.38% of the total This significant percentage of participants who have not engaged with chatbots highlights a concerning trend regarding the lack of popularity and adoption of AI chatbots among users.
The survey results indicate a gender distribution of 54% males and 46% females among participants Age demographics reveal that 8.6% are under 18, 42.9% are between 18 and 25, 33.1% fall within the 26 to 40 age range, and 15.4% are over 40 Notably, 83.7% of respondents are currently enrolled in a university The majority of participants are from the economic sector (46.3%), with a minimal representation from the service industry (0.3%) Income levels show that 31.4% earn less than 5 million VND monthly, 28.6% earn between 5 million and 10 million VND, 24% earn between 10 million and 20 million VND, and only 16% earn over 20 million VND monthly Additionally, nearly half of the respondents (43.4%) reported using the service once a month, while 34% used it twice a month, and the remaining participants used it three times or more in the past month.
Source: Suggested by the author during the research process
Table 3: Summary results of coefficients in the PLS - SEM model
Perceive the usefulness of chatbot
Ease of Use of chatbot
The reliability of the scale was assessed using Cronbach's Alpha, with all variables exceeding the minimum threshold of 0.7, confirming their significance (DeVellis, 2012) Additionally, all variables met the Composite Reliability (CR) standard of greater than 0.7 (Bagozzi & Yi, 1988) These evaluations collectively ensure that the scale's reliability is robust.
Evaluating the convergent validity of the scale: Results from Table 3 show that the total variance extracted (AVE) coefficient of the variables is > 0.5 (Hock & Ringle,
2010) and meets the testing requirements Therefore, all scales have convergent validity, so there is no need to remove any scales.
Table 4: Discriminant value according to the method using the matrix table
IN IQ LT PEI PEUxLT
Table 5: Discriminant value according to the HTMT index method
Index IN IQ LT PEI PEUxLT
Evaluating discriminant value: To evaluate the differences between the constructs in the model, we use two methods simultaneously:
According to Fornell and Larcker (Table 4), the square root of the Average Variance Extracted (AVE) values—0.957, 0.922, 0.914, 0.861, 1.000, 0.903, 0.918, 1.000, 0.912, and 0.930—exceeds the corresponding correlation values between the latent variables This indicates that the independence among the structures is maintained, confirming the robustness of the model as proposed by Fornell and Larcker (1981).
The Heterotrait-Monotrait ratio (HTMT) results, as shown in Table 5, indicate that all variables meet the strict assessment threshold of less than 0.85, as recommended by Kline (2015) This confirms the effective discrimination of the variables involved.
IN IQ LT PEU PEUxLT
The analysis presented in Table 6 indicates that the Variance Inflation Factor (VIF) for all variables is below 3, suggesting that multicollinearity is not a concern among the potential variables.
Table 7: Estimated results through the PLS - SEM model named SmartPLS
Source: Compiled by the author during the research process
* The results show the following relations:
Perceived ease of use (PEU) -> perceived usefulness (PU)
Technology knowledge (LT) X perceived usefulness (PU) -> intention to use (IN)
=> Neither of them is statistically significant as their p value exceeds 0.05, which is the confidence threshold of 95%.
Most path coefficients in the analysis indicate a positive impact on the relationships studied, with the exception of the relationship between Perceived Ease of Use (PEU) and Technology Literacy (LT) concerning Intention to Use (IN), which shows a negative path coefficient of -0.067.
The analysis reveals that Information Quality (IQ) has the most significant influence on Perceived Usefulness (PU) with a coefficient of 0.286, followed by Service Quality (SeQ) at 0.215 System Quality (SyQ) contributes a lesser impact at 0.121, while Perceived Ease of Use (PEU) shows the smallest effect with a coefficient of 0.056.
- The order of impact from strong to weak on the Perceived ease of use (PEU) variable is: Service quality (SeQ) (0.259) > System quality (SyQ) (0.148) > Information quality (IQ) (0.142).
The factors influencing the Intention to Use (IN) are ranked in order of impact: Privacy and Security (PS) has the strongest effect at 0.345, followed by Perceived Usefulness (PU) at 0.194, and Perceived Ease of Use (PEU) at 0.140 Interestingly, the interaction between Perceived Ease of Use (PEU) and Technology Literacy (LT) shows a negative impact of -0.067.
Discussion
The authors aim to bridge the research gap by enabling readers, especially banks, to comprehend the key factors that enhance the performance of Chatbots in boosting customer satisfaction To address potential biases associated with traditional analytical methods, as highlighted by Harman (1967), and to improve upon the limitations observed in studies using SPSS software (Le X c & Nguyen T H., 2024; Skandali D., Magoutas A., & Tsourvakas G., 2023; Li c Y., Fang Y H., & Chiang Y H., 2023), this research employs the SEM structural model evaluation via SmartPLS.
3 is applied and results in the absence of linear multi-addition between latent variables
The research findings indicate that the relationship between perceived ease of use (PEU) and perceived usefulness (PU), as well as the interaction of technology literacy (LT) with perceived usefulness (PU) in influencing intention to use (IN), are not statistically significant Detailed results for each variable relationship within the study model are provided below.
The quality of information provided by chatbots significantly enhances customer experience in virtual assistant services, as highlighted by D I Sensuse et al (2019) Their model analysis reveals that information quality positively influences customer perceptions of helpfulness and ease of use A survey of banking customers who have utilized chatbots confirms that the quality of information is crucial for their continued use of this technology The study emphasizes that chatbot information must be assessed through three key variables: completeness, reliability, personalization, and ease of understanding.
The evaluation of Chatbot technology reveals that system quality has a minimal effect on customers' perceptions of usefulness and ease of use This contradicts the hypothesis, as the current level of chatbot technology in banking operations has not been optimized to enhance customer experience Chatbots struggle to address complex inquiries and execute intricate tasks, such as automatically registering for credit cards or savings accounts Furthermore, the monotonous and user-unfriendly interface design of bank chatbots contributes to customer frustration and disinterest in utilizing AI consulting services.
Service quality in chatbots is evaluated based on response time, availability, and human-like characteristics Despite their long-standing use in Vietnam's banking sector, chatbots currently fall short of customer expectations, leading to a weak impact on user satisfaction and ease of use This finding contrasts with other studies suggesting that delayed responses enhance the human-like experience of chatbots In this research, the combined effects of Recovery Time and Human-like characteristics on service quality positively influence user feelings, ease of use, and helpfulness, ultimately affecting the intent to utilize chatbots As technology evolves with advancements like AI and machine learning, chatbots are poised to emerge as a leading application alongside innovations such as fintech, blockchain, open banking, and digital banking.
Chatbots offer the significant advantage of human-like interaction, providing customers with a sense of comfort However, the current limitations in technology prevent these virtual assistants from fully bridging the gap between human consultants and machines, hindering the complete realization of this benefit.
Privacy and security are critical factors in the technology adoption model, significantly influencing customer trust and technology usage decisions Model testing indicates a strong relationship (fA2=0.35) between privacy and security impacts and the intention to use chatbots in banking operations These findings highlight that users place a high value on information security and transparency regarding personal information when engaging with chatbot technology.
In the proposed model, the quality of information provided by chatbots has minimal influence on users' perception of ease of use Instead, both service quality and system quality significantly enhance the perceived usefulness of chatbot technology in banking transactions, consultations, and financial tool usage Furthermore, the evaluation of the model indicates that information quality and service quality are critical factors that lead users to recognize chatbots as a valuable technology compared to other banking solutions.
In Vietnam, the decision-making process regarding the use of bank chatbots is not significantly influenced by customers' technology background knowledge Most users find chatbot technology easy to navigate due to their familiarity with text messaging and online information searches Despite this accessibility, chatbots currently lack the capability to handle complex tasks, such as replacing bank staff for intricate customer support procedures, which limits their effectiveness As a result, even users with minimal technological expertise can still effectively utilize chatbots for basic banking inquiries.
A recent study highlights that privacy and security are the most significant factors influencing customers' intention to use AI consulting services In contrast, the primary driver for users' intent to engage with chatbots is the concept of Refreshing Value (Ma X & Huo Y., 2023) Utilizing the AIDUA research model and CAT theory, the study explores how sensory and rational factors impact user decisions regarding Chat-GPT While Chat-GPT aims to fulfill users' information needs, the element of novelty and curiosity plays a crucial role in shaping user intent Conversely, banking chatbots not only address information requests but also execute complex banking operations, making the privacy and security of user data a paramount concern.
Recent studies, including those by Ma and Huo (2023) and Alsharhan et al (2023), reinforce the idea that chatbots with human-like traits significantly influence users' decisions to adopt this technology These findings indicate that chatbots offering human-like feedback enhance the decision-making process for users, making them a popular choice as independent variables in various research models.
The Systematic Literature Review (SLR) method, guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), reveals a scarcity of research papers incorporating regulatory variables in their models Among the most frequently utilized regulatory variables are gender, experience, age, and technology anxiety (Alsharhan A., Al-Emran M., & Shaalan K., 2023).
Previous research has often overlooked technology concerns, as advancements have integrated technology into everyday life, diminishing the significance of technological barriers for users This study reinforces this notion by highlighting that technology literacy, included as a regulatory variable, has a minimal effect on the intention to use chatbots.
This study builds on the findings of Stockholm and Uppsala (2019), emphasizing the significance of human-like communication in chatbot interactions with customers It incorporates previously identified issues, such as expected privacy risks, trust in providers, and system reliability The research addresses critical theoretical and practical aspects influencing consumer intentions to use chatbots A key challenge in this field is the rapid advancement of technology and evolving consumer preferences, as noted by Brandtzæg and Heim (2009) The focus remains on factors related to innovation and consumer preferences, including Information Quality, System Quality, Service Quality, Ease of Use, and Usefulness (Sugumar M & Chandra S., 2021).
Implication
Research highlights the significant impact of the digital era, particularly in industries adapting to modern technologies Artificial intelligence consulting services have notably enhanced customer consulting, especially within the banking sector Utilizing the reliable TAM2 model, this paper serves as a valuable reference for future developments Beyond system and service quality, it also addresses critical factors for customers, such as information security and the necessity of a foundational understanding of technology for effective service usage.
Research on the application of chatbots in the banking sector is limited, making this study valuable for future exploration in this field By utilizing established technology adoption theories like UTAUT2 (Venkatesh et al., 2012), this research extends the model to include consumer preferences regarding the human-like qualities of chatbots Additionally, it highlights the importance of privacy policies, user habits, capabilities, and performance expectations, thereby enriching the existing literature on the adoption of AI-driven consulting services.
In addition, the study highlighted the role of Useful Perception and Ease of Use Perception influencing consumers' decision to use Chatbot, fully explaining the influencing factors and development directions.
The research highlights key contributions to the banking sector by emphasizing the importance of demographics in AI consulting services, enabling banks to tailor their offerings to specific customer segments It underscores the significance of information quality indicators, as customers are more likely to continue using chatbots if they receive complete and accurate responses To enhance user experience, banks should focus on improving three critical impact variables: completeness, relevance, and personalization of information Additionally, the quality of service significantly influences customer satisfaction, suggesting that banks should prioritize factors such as response time, availability, and a human-like interaction style to attract a broader audience An intuitive interface design is also essential to engage users who may not be tech-savvy Finally, ensuring robust information security measures is crucial for instilling customer confidence in AI consulting services.
Customer surveys provide valuable insights into their perceptions of banking systems, enabling banks to enhance their AI-driven consulting services This process not only improves service quality but also ensures that customers receive the best possible experience and have the opportunity to choose from a range of AI technology consulting options offered by different banks.
The research paper highlights significant contributions to Vietnamese state agencies by emphasizing the importance of artificial intelligence as a global trend By understanding customer sentiments regarding AI-driven consulting services, banks in Vietnam can align with international advancements and embrace the digital era The adoption of new technologies not only enhances efficiency and resource management but also promotes sustainable development within the financial markets.
Conclusions and recommendations
Chatbots have gained popularity worldwide, yet their application in Vietnam's credit institutions and banks is a recent development Historically, both banks and developers moved away from this technology due to a lack of interest and understanding among Vietnamese customers However, following the COVID-19 pandemic, there has been a significant increase in users, driven by advancements in technical infrastructure and the growth of the information society As a result, many businesses and credit institutions are now focusing on and investing in AI consulting services.
Despite the challenges in enhancing online communication services to satisfy customer demands, the advancement of AI-driven consulting services presents a significant opportunity By fully leveraging AI capabilities, credit institutions and banks in Vietnam can achieve substantial improvements in their service offerings, ultimately benefiting the broader field of service-related consulting.
To enhance customer service with human-like interactions, banks and fintech companies must invest significant time and financial resources into chatbot technology While these initial costs may not yield immediate profits, such investments align with government digital transformation initiatives and could pave the way for future advancements in the rapidly evolving AI landscape This research aims to identify the factors influencing customer experience with chatbot applications in transaction and consulting services at commercial banks in Vietnam, enabling technology developers to better understand customer expectations and trends in AI consulting.
Through the results of the study, to improve the performance and attractiveness of
AI consulting technology in the banking sector in Vietnam, the authors have the following suggestions:
Ease of use is a crucial factor driving customers to adopt digital communication, prompting banks to create transaction systems with user-friendly interfaces and clear, understandable information In Vietnam, the government aims to transition daily banking operations to digital platforms; however, varying levels of digital literacy across different regions pose challenges Therefore, financial institutions must ensure that their instructional materials are accessible to individuals of all educational backgrounds, enabling new customers to quickly learn and effectively utilize chatbot applications and other digital tools.
To enhance customer security and privacy, it is crucial for banks and financial institutions to prioritize the development of robust security systems in their digital services By ensuring that customers feel safe while utilizing new technologies to address their needs, these institutions can foster trust in their digital offerings As traditional consultants face declining trust due to concerns over personal information sales and unauthorized data usage, virtual consultants have a unique opportunity to capitalize on this shift by investing in advanced security measures for banking applications.
To demonstrate the effectiveness of chatbots, banks and financial institutions must prioritize the collection of complete, truthful, and accurate information, ensuring data reliability By leveraging AI, these institutions can personalize user profiles to deliver tailored responses that address specific customer inquiries and preferences Additionally, the enhancement of response speed allows for continuous service availability, enabling AI to provide consulting assistance 24/7, unlike traditional human support which is time-restricted.
7.3 Limitations and further research directions
Future research must address certain limitations, including the unvalidated information provided by participants, which introduces bias in the data Additionally, the survey's sample size is insufficient to ensure accuracy across Vietnam, as most responses are predominantly from residents of Ho Chi Minh City Consequently, the findings do not reflect a comprehensive evaluation of the topic.
The research primarily examines text-based chatbot technology, yet it overlooks conversational chatbot capabilities This gap is significant as it may impact user experience and the intention to utilize AI consulting services in the banking sector.
Research by Yu and Zhao (2024) highlights the significance of visual elements, particularly emojis, in enhancing user satisfaction with chatbot service quality Their findings reveal that the use of emojis fosters a sense of warmth, which plays a crucial role in mediating satisfaction levels Notably, the warmth conveyed through emojis has a more pronounced effect on user satisfaction in pre-programmed chatbots compared to AI-powered chatbots.
The research paper overlooks the influence of current advanced artificial intelligence technologies, including Chat-GPT, Bing, and Monica, on customers' intentions to adopt chatbot technology It is essential to explore whether these technologies serve as significant variables affecting consumer choices, and to determine the nature of this impact—whether it is direct or indirect, aligned or contrary, and its strength.
The paper notably overlooks the needs of users with disabilities, highlighting the necessity for AI-based consulting services to enhance support for these individuals in financial transactions within banking It is crucial for technology to cater to all customers, as its effectiveness is measured by the breadth of users it serves and their satisfaction Consequently, the development of AI consulting services aimed at assisting people with disabilities in banking operations is an essential area for future research and improvement.
Alagarsamy, s., & Mehrolia, s (2023) Exploring chatbot trust: Antecedents and behavioural outcomes Heliyon, 9(5). https://doi.org/10.1016/j heli yon.2023 e 16074
Alsharhan, A., Al-Emran, M., & Shaalan, K (2023) Chatbot Adoption: A
Multiperspective Systematic Review and Future Research Agenda IEEE
TRANSACTIONS ON ENGINEERING MANAGEMENT. http://dx.doi.org/10.1109/TEM.2023.3298360
Brandtzaeg, p B., & Folstad, A (2017) Why People Use Chatbots Internet
David (1989) Perceived Usefulness, Perceived Ease of Use, and User
Acceptance of Information Technology MIS Quarterly, 13(3), 319-340 https://doi.org/10.2307/2490Q8 '
Heo, M., & Lee, K J (2018) Chatbot as a New Business Communication
Tool: The Case of Navcr TalkTalk Business Communication Research and
Practice, (1(1)), 41-45 https://doi.org/10.22682/bcrp.2Ql 8.1 ■ 1.41
Jennerboer, L., Herrando, c., & Constantinides, E (2022) The Impact of
Chatbots on Customer Loyalty: A Systematic Literature Review Journal of
Theoretical and Applied Electronic Commerce Research, I7(\), 212-229 https://doi.Org/l 0.3390/jtaerl 7010011
Laumer, s., Maier, c., & Gubler, F T (2019) CHATBOT ACCEPTANCE IN HEALTHCARE: EXPLAINING USER ADOPTION OF
CONVERSATIONAL AGENTS FOR DISEASE DIAGNOSIS European
Conference on Information Systems, 27 https://aisel.aisnet.org/ecis2019 rp/88
Le, c X., & Tran, N H (2024) The effects of chatbot characteristics and customer experience on satisfaction and continuance intention toward banking chatbots: Data from Vietnam Data in Brief, 52. https://d0i.0rg/l 0,1016/j di b 2023.110025
Li, C.-Y., Fang, Y.-H., & Chiang, Y.-H (2023) Can AI chatbots help retain customers? An integrative perspective using affordance theory and service domain logic Technological Forecasting and Social Change, 197 https://doi.org/10.1016/j tech fore.2023.122921
Li, C.-Y., & Yang, J.-T (2023) Chatbots or me? Consumers' switching between human agents and conversational agents Journal of Retailing and
Consumer Services, 72, 103264. https://d0i.0rg/l 0.1016/j.jretconser.2023.103264
A recent study by Liu et al (2023) investigates user satisfaction with task-oriented chatbots in mainland China and Hong Kong Utilizing a mixed-method approach, the research highlights key differences in user experiences across these regions The findings reveal that cultural and contextual factors significantly influence satisfaction levels, emphasizing the importance of tailoring chatbot interactions to meet diverse user needs This study contributes valuable insights into the design and implementation of chatbots, aiming to enhance user engagement and effectiveness in various settings.
Ly, B., & Ly, R (2022) Internet banking adoption under technology acceptance model—Evidence from Cambodian users Computers in Human
Behavior Reports, 7, 100224 https://doi.org/10.1016/j.chbr.2O22.100224
Ma, X., & Huo, Y (2023) Are users willing to embrace ChatGPT? Exploring the factors on the acceptance of chatbots from the perspective of AIDUA framework Technology in Society, 75 https://doi.Org/10.1016/j.techsoc.2023.102362
Min, s., So, K K F., & Jeong, M (2018) Consumer adoption of the Uber mobile application: Insights from diffusion of innovation theory and technology acceptance model Journal of Travel & Tourism Marketing, 36(7), 770-783 https://doi.Org/l 0,1080/10548408.2018.1507866
Ngân hàng Nhà nước đã ban hành Quyết định 810/QĐ-NHNN vào năm 2021, phê duyệt Kế hoạch Chuyển đổi số ngành Ngân hàng đến năm 2025 Quyết định này nhằm thúc đẩy quá trình chuyển đổi số trong lĩnh vực ngân hàng, nâng cao hiệu quả hoạt động và đáp ứng nhu cầu ngày càng cao của khách hàng Thông tin chi tiết có thể được tìm thấy tại Thư viện pháp luật qua liên kết: https://thuvienphapluat.vn/van-ban/Tien-te-Ngan-hang/Quyet-dinh-810-QD-NHNN-2021-phe-duyet-Ke-hoach-Chuyen-doi-so-nganh-Ngan-hang-den-2025-474917.aspx.
Nguyen, T., & Dinh, T (2023) Nghiên cứu các yếu to tác động đen sự chấp nhận Chatbot AI cua khách hàng tại một sỏ ngán hàng thương mại Tạp chi
The study investigates the factors influencing customer acceptance of AI chatbots in various banks, highlighting the role of user experience and trust It emphasizes the importance of mobile-based assessments in enhancing user interaction and satisfaction with banking services By analyzing customer feedback, the research aims to improve the implementation of AI technologies in the banking sector, ultimately fostering a more efficient and responsive customer service environment.
Investigating the factors that influence behavioral intention to use Computers
& Education, https://doi.org/10.1016/j.compcdu.2017.02.005
Pega (2019) Fusing AI with empathy Pega https://www.pega.com/fusing-ai- with-empathy
Petersson, A H., Pawar, s., & Fagerstrpm, A (2023) Investigating the factors of customer experiences using real-life text-based banking chatbot: a qualitative study in Norway Procedia Computer Science, 219, 697 - 704. https://doi.org/10.1016/j procs.2023.01.341
Przegalinska, A., Ciechanowski, L., Stroz, A., Gloor, p., & Mazurek, G (2019)
In bot we trust: A new methodology of chatbot performance measures Business
Horizons, 62(6), 785-797 https://doi.Org/10.1016/j.bushor.2019.08.005
Rahman, M., Ming, T H., Baigh, T A., & Sarker, M (2018) Adoption of artificial intelligence in banking services: an empirical analysis International
Journal of Emerging Markets, https://doi.org/10.1108/IJOEM-06-2020-0724
Skandali, D., Magoutas, A., & Tsourvakas, G (2013) Artificial Intelligent
Applications in Enabled Banking Services: The Next Frontier of Customer
Engagement in the Era of ChatGPT Theoretical Economics Letters, 13, 1203
Sugumar, M., & Chandra, s (2021) o 1 Desir Do I Desire Chatbots t e
Chatbots to be lik o be like Humans? Exploring F e Humans? Exploring Factors for ors for Adoption of Chatbots for Financial Services Journal of
International Technology and Information Management, 30(3) https://doi.org/10.5 8729/1941-6679.1501
Quyết định 986/QĐ-TTg năm 2018 được Thủ tướng Chính phủ phê duyệt nhằm phát triển ngành ngân hàng Việt Nam giai đoạn 2025-2030 Chiến lược này tập trung vào việc nâng cao hiệu quả hoạt động, tăng cường quản lý và phát triển công nghệ ngân hàng, đồng thời đảm bảo an toàn và ổn định cho hệ thống tài chính Thông tin chi tiết có thể tham khảo tại Thư viện pháp luật qua liên kết: https://thuvienphapluat.vn/van-ban/Tien-te-Ngan-hang/Quyet-dinh-986-QD-TTg-2018-phe-duyet-Chien-luoc-phat-trien-nganh-Ngan-hang-Viet-Nam-2025-2030-390316.aspx.