Enhancing Financial Advisory Services with GenAI: Consumer Perceptions and Attitudes Through Service-Dominant Logic and Artificial Intelligence Device Use Acceptance Perspectives
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsEnhancing Financial Advisory Services with GenAI: Consumer Perceptions and Attitudes through Service-Dominant Logic and Artificial Intelligence Device Use Acceptance Perspectives
This is an exciting paper. However there are issues that need to be addressed to improve the paper.
1. The abstract is well-written, presented in a clear and logical manner.
2. The research question has been clearly stated in the introduction, but it needs to be directed towards the novelty of the study to make the state of the art more evident. It would be better if the novelty originated from the suggestions for future research provided in previous studies.
3. Please provide the structure of the article in the last paragraph of the introduction section.
4. The literature review and hypotheses are clearly developed based on the gaps and results of previous studies.
5. Figure 1. Research Model should include hypothesis numbers. Please consider this article as one of the references https://doi.org/10.1002/bse.3393
6. Purposive sampling targeting adult mobile banking users’ needs a more detailed explanation, such as the respondents' age range, the data collection period, the platform used for data collection, and the total number of respondents before arriving at the final 822 samples. Moreover, Table 1. Demographic statistics shows respondents' ages not only within the adult range but also above 65 years.
7. Information about the pilot test also needs to be clarified, such as the sample size used in the pilot survey. Mention or explain if any variables or indicators were revised during the pilot survey. Also, provide references related to the pilot survey process.
8. The author(s) need to standardize the number of decimal places after a point or comma. Use three decimal places in Table 2: Reliability, CR, and AVE. Apply this standardization throughout the paper.
9. Typical discussion is missing, it is crucial part of the manuscript, compare your results to similar studies.
10. Please check the spelling of terms, for example, Human-like Enpathy (HLE)
Best of luck, and I hope this comment is helpful.
Author Response
Dear Reviewer,
Thank you for your thoughtful and constructive feedback on our manuscript titled "Enhancing Financial Advisory Services with GenAI: Consumer Perceptions and Attitudes through SDL and AIDUA Perspectives." We have carefully considered all of your suggestions and made the necessary revisions to improve the paper. Please find below our responses to your comments, with details of the changes made in the manuscript.
Comments 1: The abstract is well-written, presented in a clear and logical manner.
Response 1: Thank you for your positive feedback on the abstract. We appreciate your comments and are pleased that the abstract has met your expectations. No further changes were made to this section.
Comments 2: The research question has been clearly stated in the introduction, but it needs to be directed towards the novelty of the study to make the state of the art more evident. It would be better if the novelty originated from the suggestions for future research provided in previous studies.
Response 2: Thank you for your valuable feedback. We agree with your suggestion to more clearly direct the research question towards the novelty of the study and align it with the suggestions for future research provided in previous studies. In response to your comment, we have revised the introduction to emphasize the research gap identified in prior studies, particularly concerning the impact of GenAI on user interactions in financial services. By referencing the need for further exploration of GenAI’s unique attributes, as highlighted by prior research, we position the study’s contribution more clearly within the existing literature.
The updated text reads as follows:
Before: "However, the introduction of GenAI has further complicated user interactions, potentially influencing consumers' attitudes and responses to these services [6]."
After: "However, these studies have largely overlooked the specific influence of GenAI technologies, particularly in terms of how their distinct attributes reshape user experiences in financial contexts. The gap identified by prior research [6] suggests a need for future studies to explore how GenAI technologies, with their conversational nature and capacity for continuous learning, influence consumer perceptions of financial advice services."
Before: "This study suggests that these characteristics and their influence on consumer responses can be analyzed using service-dominant logic (SDL) and AI Device Use Acceptance (AIDUA) frameworks [8,9]."
After: "Building on the gaps identified in earlier studies, we integrate service-dominant logic (SDL) and AI Device Use Acceptance (AIDUA) frameworks to explore the role these attributes play in shaping consumer trust and acceptance of GenAI-based financial advisory services [8,9]."
Comments 3: Please provide the structure of the article in the last paragraph of the introduction section.
Response 3: Thank you for the suggestion. We have now added a summary of the article’s structure at the end of the introduction. This addition provides readers with an overview of the paper’s organization, ensuring better readability and clarity.
The updated text reads as follows:
"This paper is organized as follows: Section 2 presents a literature review and the theoretical framework, focusing on the evolution of financial advisory services and the unique attributes of GenAI. Section 3 develops the research hypotheses and model, integrating service-dominant logic (SDL) and AI Device Use Acceptance (AIDUA) frameworks. Section 4 outlines the research methodology, including data collection and measurement development. Section 5 discusses the data analysis and results of the structural model. Finally, Section 6 provides the conclusion, academic and practical implications, and suggestions for future research directions."
Comments 4: The literature review and hypotheses are clearly developed based on the gaps and results of previous studies.
Response 4: Thank you for your positive feedback on the abstract. We appreciate your comments and are pleased that the literature review and hypotheses have met your expectations. No further changes were made to this section.
Comments 5: Figure 1. Research Model should include hypothesis numbers.
Response 5: Thank you for your comments. We added hypothesis numbers in Figure 1.
Comments 6: Purposive sampling targeting adult mobile banking users’ needs a more detailed explanation, such as the respondents' age range, the data collection period, the platform used for data collection, and the total number of respondents before arriving at the final 822 samples. Moreover, Table 1. Demographic statistics shows respondents' ages not only within the adult range but also above 65 years.
Response 6: Thank you for the insightful comment. We have revised the data collection section to provide more details about the sampling approach, including the age range of respondents (18 to above 65 years), the total number of respondents (1,200 invited, 950 completed, 822 retained), and the platforms used for data collection (email campaigns, in-app notifications, and financial forums). The data collection period (January to March 2024) has also been specified. These revisions provide a clearer and more comprehensive explanation of the sampling method and respondent characteristics.
The updated text reads as follows:
Before: The initial description of the sampling method and respondents mentioned targeting adult mobile banking users but did not provide specific details on the total number of respondents, the data collection platforms, or the presence of respondents above 65 years.
After: The revised section now includes details such as the total number of respondents invited (1,200), the final number of valid responses (822), and the inclusion of older respondents (above 65 years). It also specifies the data collection period (January to March 2024) and the platforms used for survey distribution (email campaigns, in-app notifications, financial forums, and social media). Additionally, it explains that a pilot test was conducted before the formal survey launch.
Comments 7: Information about the pilot test also needs to be clarified, such as the sample size used in the pilot survey. Mention or explain if any variables or indicators were revised during the pilot survey. Also, provide references related to the pilot survey process.
Response 7: Thank you for your helpful feedback. I/We have revised the pilot test section to include specific details about the sample size (50 participants) and the variables revised during the test, such as AI literacy, perceived authenticity, and human-like empathy. Additionally, I/We have added a reference to Van Teijlingen & Hundley (2001) to provide context and justification for the pilot survey process. These revisions clarify how the pilot test contributed to refining the survey instrument.
The updated text reads as follows:
"Before launching the formal survey, a pilot test was conducted with a subset of 50 participants to identify and address any potential issues with clarity, question comprehensibility, and the overall structure of the questionnaire. The pilot survey helped refine variables such as AI literacy, perceived authenticity, and human-like empathy, following the recommendations of Van Teijlingen & Hundley (2001)."
Comments 8: The author(s) need to standardize the number of decimal places after a point or comma. Use three decimal places in Table 2: Reliability, CR, and AVE. Apply this standardization throughout the paper.
Response 8: Thank you for your feedback. We standardized the number of decimal places after a point or comma.
Comments 9: Typical discussion is missing, it is crucial part of the manuscript, compare your results to similar studies.
Response 9: Thank you for your valuable feedback. We agree with your comment regarding the need for a more comprehensive discussion, particularly a comparison of our findings with those from similar studies. In response to your comment, We have revised the Academic Implications section to provide a clearer discussion, drawing comparisons with previous research where relevant. We have also added references to similar studies to highlight how our findings align with, or extend, existing knowledge in the field.
The updated discussion emphasizes:
The significance of personalized investment suggestions, human-like empathy, and continuous improvement in co-creating value, aligning with Service-Dominant Logic (SDL) principles (Ref. [90]).
The role of human-like empathy in influencing perceived authenticity, contributing to literature on AI technology design (Ref. [91]).
The correlation between perceived authenticity and utilitarian attitudes, expanding on technology adoption theories and highlighting the importance of authenticity in AI acceptance (Ref. [96]).
The importance of AI literacy in enhancing the impact of GenAI's features on perceived authenticity and willingness to communicate (Ref. [97]).
Comments 10: Please check the spelling of terms, for example, Human-like Enpathy (HLE)
Response 10: Thank you for your feedback. We checked the spelling of terms and revised all.
Reviewer 2 Report
Comments and Suggestions for AuthorsThis paper reads more like a summary of a consumer survey a bank would put out than a scientific research paper in finance. There is no formal scientific rigor on the survey questions, nor the specific theory the paper was trying to address. Here are two major issues I have with the design of the study.
1. I cannot get a clear picture of what the GAI services are being offered. The only item that could count as a service is personalized investment advice. This alone could be a topic for rigorous testing, with the appropriate survey design. Other so-called services are more like attributes of the GAI. For example, H1 states that personalized investment advice (suggestion) is positively associated with perceived authenticity. What personalized investment advice is being offered? Who determines what constitutes perceived authenticity? The three questions in the survey related to PA are not good enough as the sources are not scientific papers.
2. All the hypotheses are based on questionable foundations. For example, H9 states that "Consumers' AI literacy positively moderates the relationship between GenAI's human-like empathy and consumers' perceived authenticity. " What does "positively moderates" mean? If a GAI shows human-like empathy, it has to be seen as more authentic. But who determines if a GAI displays human-like empathy? By definition, all GAI models should display human-like empathy as they're designed to be more and more like humans.
I wonder if either of the authors truly understands what a GAI is. This paper reads too much like something a GAI would write: reads good on the surface and lacks rigor in the content. The author should consider submitting the paper to consumer preference journal that don't demand as much scientific rigor as a finance journal would.
The paper reads too much like a GAI-generated draft. While grammatically correct and with no spelling errors, it lacks "authenticity".
Author Response
Dear Reviewer,
Thank you for your thorough and insightful feedback on our manuscript. We appreciate the opportunity to clarify the points you raised and to improve the overall quality of the study. Below is a detailed response to your comments and concerns.
Comments 1: I cannot get a clear picture of what the GAI services are being offered. The only item that could count as a service is personalized investment advice. This alone could be a topic for rigorous testing, with the appropriate survey design. Other so-called services are more like attributes of the GAI. For example, H1 states that personalized investment advice (suggestion) is positively associated with perceived authenticity. What personalized investment advice is being offered? Who determines what constitutes perceived authenticity? The three questions in the survey related to PA are not good enough as the sources are not scientific papers.
Response 1: Thank you for raising these important points. We agree that greater clarity is needed regarding the nature of the GenAI (Generative AI) services being examined. Below, we address each aspect of your comment in detail.
1) Clarification of GenAI Services
The GenAI services studied in this paper revolve around its core ability to offer personalized financial advice, particularly in the form of investment suggestions. These services go beyond basic algorithmic advice (as offered by traditional robo-advisors) by employing advanced conversational AI to simulate human-like interactions and provide personalized, adaptive investment recommendations based on real-time market data and user preferences. The unique attributes of GenAI, such as human-like empathy and continuous improvement, complement these services by enhancing the overall user experience and perceived authenticity of the advice.
Personalized investment suggestions, as conceptualized in this study, include a series of tailored recommendations that take into account a user’s financial goals, risk tolerance, past investment behavior, and market conditions. These suggestions are dynamic and responsive to changing circumstances, which sets them apart from the standardized advice typically provided by robo-advisors.
2) Perceived Authenticity (PA)
We acknowledge your concern about the conceptualization and measurement of perceived authenticity (PA). The concept of perceived authenticity in AI-driven services is a relatively new area of research, and while we used sources outside traditional finance literature, our approach was inspired by literature in human-computer interaction (HCI) and AI ethics, where authenticity has been defined in terms of truthfulness, reliability, and the perceived genuineness of interactions between users and technology. We agree that more rigor could be applied to linking this concept to finance-specific studies, but if we consider your comments, we may have to rewrite the paper in its entirety. Therefore, we hope to fully address your comments in subsequent studies.
3) Survey Design and Question Validity
We appreciate your feedback regarding the survey questions related to PA. To improve the scientific rigor, we have to revisit the design of the survey questions to ensure they are directly tied to validated constructs in the literature. However, if we consider your comments, we may have to rewrite the paper in its entirety. Therefore, we hope to fully address your comments in subsequent studies.
Comments 2: All the hypotheses are based on questionable foundations. For example, H9 states that "Consumers' AI literacy positively moderates the relationship between GenAI's human-like empathy and consumers' perceived authenticity. " What does "positively moderates" mean? If a GAI shows human-like empathy, it has to be seen as more authentic. But who determines if a GAI displays human-like empathy? By definition, all GAI models should display human-like empathy as they're designed to be more and more like humans.
Response 2: We appreciate your concerns regarding the clarity of the hypotheses, particularly the moderating role of AI literacy in Hypothesis 9 (H9). Below, we explain the theoretical rationale behind this hypothesis and the interpretation of moderation in the context of social science research.
1) Clarification of "Positively Moderates"
In social science research, moderation refers to the interaction effect of a third variable on the relationship between two other variables. In this context, when we state that "AI literacy positively moderates the relationship between GenAI’s human-like empathy and consumers’ perceived authenticity," we mean that the effect of human-like empathy on perceived authenticity becomes stronger for consumers who have higher levels of AI literacy. Essentially, AI-literate consumers are more likely to recognize and appreciate the human-like empathetic traits of GenAI, leading to a greater perception of authenticity.
Moderation does not imply that GenAI models always perfectly exhibit empathy, nor does it assume that all users will perceive human-like empathy to the same degree. The moderating effect of AI literacy posits that consumers with a better understanding of AI’s capabilities are more capable of identifying and interpreting empathetic behavior in AI systems, which may enhance their overall perception of authenticity. In contrast, those with lower AI literacy might not fully recognize these nuances and may not perceive the AI’s responses as authentically empathetic, even when they are designed to be so.
Moderation effects are a common and well-established approach in behavioral science and marketing research, particularly when examining the interaction between consumer characteristics (like AI literacy) and their reactions to technological attributes (like human-like empathy). This interaction is important because individual differences—such as familiarity with technology—can influence how consumers interpret and respond to AI-driven services.
2) The Role of AI Literacy
The concept of AI literacy is key to understanding Hypothesis 9. AI literacy refers to consumers’ knowledge and familiarity with how AI works, including its strengths and limitations. Our hypothesis is based on the idea that consumers with higher AI literacy are better equipped to detect when an AI system is effectively mimicking human-like empathy. For these consumers, the human-like interactions of GenAI are more likely to be perceived as genuine and trustworthy, thus increasing the perceived authenticity of the service. This is supported by literature that suggests cognitive ability and domain knowledge can shape how users evaluate new technologies (e.g., McKnight et al., 2002; Pavlou & Gefen, 2004).
On the other hand, consumers with lower levels of AI literacy may not fully comprehend the nuances of GenAI’s empathetic behaviors. These users may struggle to distinguish between mechanical or automated responses and genuinely empathetic interactions, which could dampen their perception of authenticity. Thus, the impact of human-like empathy on perceived authenticity is likely to be weaker among consumers who lack a clear understanding of AI's capabilities.
Perception of Human-Like Empathy: We agree with your point that human-like empathy is a designed attribute of GenAI systems, and many such systems are developed to simulate empathy. However, there is variability in how well different AI systems can mimic empathy, and more importantly, there is variability in how different users perceive that empathy. Just because an AI is designed to be empathetic does not guarantee that all users will perceive it as such.
Perceived empathy is subjective, and it can be influenced by users' prior experiences with AI, their expectations, and their understanding of how AI works. Therefore, it is crucial to investigate how individual differences—such as AI literacy—affect the interpretation of these empathetic behaviors. The fact that AI is designed to be empathetic does not mean that every user will experience it as such, especially if they lack familiarity with the underlying mechanisms of AI systems. This is why we believe that examining moderating effects is not only appropriate but necessary in understanding diverse consumer responses to AI-driven services.
Scientific Basis for Moderation Hypotheses: Moderation effects are a well-established methodological approach in the social sciences, particularly in studies exploring the interaction between technology and user characteristics. Many previous studies have applied moderation hypotheses to examine how individual traits (such as digital literacy or cognitive ability) influence the impact of technological features on user outcomes. For example, Pavlou and Gefen (2004) explored how trust moderates the effect of perceived ease of use on user intentions, while McKnight et al. (2002) examined how trust moderates the relationship between perceived website quality and usage behavior.
Our hypothesis follows this tradition by proposing that AI literacy moderates the impact of GenAI's human-like empathy on users' perceived authenticity. It is not simply assumed that all GenAI systems perfectly exhibit empathy or that all users will perceive it equally. The moderating role of AI literacy adds a nuanced understanding of how individual differences affect user experiences with AI-driven services.
Comments 3: I wonder if either of the authors truly understands what a GAI is. This paper reads too much like something a GAI would write: reads good on the surface and lacks rigor in the content. The author should consider submitting the paper to consumer preference journal that don't demand as much scientific rigor as a finance journal would.
Response 3: We understand that the presentation of Generative AI (GAI) in our paper might have prompted some questions about the depth of the content. We assure you that we have a comprehensive understanding of GAI and its applications in the financial sector. GAI refers to AI models capable of generating new data, outputs, or insights by mimicking human-like reasoning and interaction. These models are underpinned by deep learning and natural language processing technologies, allowing them to simulate human conversation and behavior, making them highly relevant in fields like financial advisory services, where personalized advice and user interaction are key.
Our paper focuses on how these AI-driven models, particularly GAI, are transforming financial advisory services by delivering personalized financial advice, interacting with users in a human-like manner, and continuously learning from user data. These aspects of GAI are highly relevant to finance journals because they explore the evolving dynamics of AI in the financial services industry, where technology is increasingly replacing traditional human roles in advisory capacities.
1) Relevance to Finance Journals
We respectfully disagree with the suggestion that the paper would be better suited for a consumer preference journal. The core of our study is not about basic consumer preferences but rather about the financial implications of GAI technologies in providing personalized financial advice, simulating human empathy, and continuously adapting to financial market conditions. The findings of this study have clear implications for the financial services industry, especially as financial institutions increasingly adopt AI-driven solutions to improve their advisory services and engage with a broader range of clients.
Financial decision-making, customer trust, and technology adoption in financial services are critical areas of research that intersect with the adoption of GAI technologies. As a result, our paper contributes directly to discussions in the field of finance about how AI systems are shaping consumer behavior and financial outcomes.
2) Scientific Rigor
While we acknowledge that the manuscript may need some refinement in how we convey the technical depth of GAI, we believe the current approach does not undermine the scientific rigor of the study. The research employs established theoretical frameworks such as Service-Dominant Logic (SDL) and AI Device Use Acceptance (AIDUA) to examine how specific attributes of GAI, such as personalized investment advice, human-like empathy, and continuous learning, impact consumer trust and decision-making. These concepts are rooted in behavioral finance, technology adoption, and service innovation, making the study well-suited for a finance audience.
Moreover, the study’s empirical design, which utilizes structural equation modeling (SEM) to analyze data from a large sample of 822 mobile banking users, adheres to the methodological standards expected in finance research. Our study extends the current literature by providing a unique examination of AI literacy and its moderating effects on the relationship between GAI’s human-like empathy and perceived authenticity—an emerging area in AI and financial services.
3) GAI and Financial Applications
We also want to emphasize that while GAI may be designed to simulate human-like empathy, not all GAI systems achieve this equally, nor do all users perceive this empathy in the same way. This is why our study delves into the role of AI literacy as a moderator in understanding how different users perceive GAI’s empathetic behavior. The variability in user responses to GAI’s empathy is particularly relevant in financial services, where trust, authenticity, and the perception of personalized care significantly affect financial decisions and outcomes.
As financial services increasingly incorporate GAI technologies to interact with customers and provide advisory services, it is crucial to understand how these systems are perceived by consumers and how different attributes, such as empathy and personalized advice, impact their trust in the financial advice provided. Our paper contributes to this discussion by examining these relationships in a finance-specific context, rather than a general consumer preference context.
We believe our paper contributes to the ongoing conversation about the integration of AI technologies in finance and provides valuable insights into the role of Generative AI in transforming financial advisory services. We are confident that the current scope of the paper aligns with the expectations of a finance journal, particularly in its focus on technology adoption, consumer trust, and financial decision-making processes.
Thank you for your thoughtful feedback. We hope this response clarifies our understanding of GAI and the relevance of our study to the finance field. We welcome any further discussion or suggestions.
Comments 4: The paper reads too much like a GAI-generated draft. While grammatically correct and with no spelling errors, it lacks "authenticity".
Response 4: Thank you for your feedback regarding the quality of the English language in our manuscript. We appreciate your observation about the language and the impression that it may read too much like a GAI-generated draft.
We want to assure you that the manuscript has undergone extensive proofreading and editing to ensure grammatical accuracy and clarity. We carefully reviewed the language to meet academic standards while ensuring that the content is both precise and readable. Our aim was to maintain a balance between technical accuracy and readability for a broad academic audience.
While we acknowledge that the writing may have come across as somewhat formal or mechanical, we have taken care to craft each section to clearly convey our arguments and research findings. We recognize the importance of “authenticity” in academic writing and have strived to ensure that our voice as authors is reflected in the manuscript. The style we employed was intended to prioritize clarity and rigor, given the complexity of the topic.
Nevertheless, we greatly value your feedback and are committed to further improving the manuscript. We will revisit the text once more to ensure it strikes the right balance between professionalism, clarity, and an engaging tone. Our team is dedicated to presenting this research in the most authentic and compelling manner possible, while maintaining high academic standards.
Thank you again for your thoughtful comments, and we look forward to further improving the manuscript.
Reviewer 3 Report
Comments and Suggestions for AuthorsChapter 2.7 Integrating SDL and AIDUA
Please consider strengthening the rationale for integratin these models by adding more evidence and literature-based arguments. Including additional references and citing previous research on best practices for combining these modles would provide a more solid foundation for your methodological approach.
- recheck the text for typos (i.e. Figure 1 - LiterCY...)
Author Response
Comments 1: Chapter 2.7 Integrating SDL and AIDUA
Please consider strengthening the rationale for integratin these models by adding more evidence and literature-based arguments. Including additional references and citing previous research on best practices for combining these modles would provide a more solid foundation for your methodological approach.
Response 1: Thank you for this valuable suggestion. We agree that providing more robust theoretical support for the integration of SDL and AIDUA would enhance the clarity and rigor of our methodological approach. In response to your comment, we have revised Chapter 2.7 by incorporating additional literature-based arguments and evidence from recent studies on value co-creation and AI adoption.
We now explicitly highlight how SDL’s value co-creation perspective complements the AIDUA model’s focus on consumer interaction stages, particularly in AI-driven services like GenAI. This integration is further supported by references to recent research that demonstrates the effectiveness of combining service logic and consumer technology adoption frameworks, particularly in financial services. The revised section also cites studies that emphasize the role of AI technologies in enhancing consumer engagement and co-creating value, aligning with the SDL perspective.
The revised Chapter 2.7 now includes additional references to key literature, including:
Vargo et al. (2008) and Grönroos (2008) on value co-creation in service logic;
Riikkinen et al. (2018) and Pelau et al. (2021) on AI technologies fostering co-creation and perceived anthropomorphism;
Bag et al. (2022) and Vesanen (2007) on AI personalization and consumer engagement in digital service environments.
These additions provide a stronger theoretical foundation and more robust support for the integration of SDL and AIDUA. The revised content can be found in Chapter 2.7 of the manuscript.
The updated text reads as follows:
The seamless integration of Service-Dominant Logic (SDL) and the AI Device Use Acceptance (AIDUA) model provides a comprehensive theoretical foundation for under-standing and explaining consumer interactions with Generative AI (GenAI) in the service industry, particularly within financial services. By combining SDL’s value co-creation perspective with AIDUA’s focus on consumers’ appraisal stages of AI usage, we create a powerful framework for investigating the nuances of consumer interactions with GenAI.
SDL emphasizes value co-creation through interaction and resource integration between service providers and consumers, aligning closely with the AIDUA model, which highlights consumer acceptance and resistance toward AI technologies. The two frameworks converge in the context of value-driven usage of AI, where consumers are not passive recipients but active participants in the co-creation of value [42,43]. Previous studies have shown that AI technologies, when effectively integrated into service systems, enhance the consumer’s role in co-creating personalized value, resulting in higher engagement and satisfaction [44].
This framework posits that, when services are designed to facilitate consumers' active role in co-creating personalized value (a fundamental concept of SDL), their experiences with AI-driven systems, like GenAI, could be significantly enhanced. AIDUA comple-ments this by focusing on the stages of consumers’ AI interaction, from initial awareness to full acceptance, which includes their evaluation of perceived authenticity, personaliza-tion, and continuous improvement—factors central to AI-human collaboration [48,50]. Furthermore, evidence suggests that consumers’ willingness to embrace AI in service set-tings increases when AI systems exhibit characteristics such as empathy and anthropo-morphism, which can foster more authentic and engaging interactions [7,49].
The decision to integrate SDL and AIDUA is also supported by recent research in both the service and AI literature. For example, studies have highlighted the effectiveness of combining consumer technology adoption frameworks with service logic to explain the adoption of AI-driven services, particularly in high-involvement contexts like financial services [9,50]. By integrating these models, we offer a more holistic understanding of how consumers perceive and engage with AI-based financial advisory services.
Comment 2: recheck the text for typos (i.e. Figure 1 - LiterCY...)
Response 2: Thank you for your feedback. We have corrected all typos through proofreading.
Reviewer 4 Report
Comments and Suggestions for AuthorsThis paper explores the impact of generative AI attributes on consumer perceptions and attitudes towards financial advisory services. It employs Service-Dominant Logic and Artificial Intelligence Device Use Acceptance frameworks to analyze data from 822 mobile banking users. The study reveals that personalized investment suggestions, human-like empathy, and continuous improvement of Gen AI positively influence perceived authenticity, which in turn fosters a utilitarian attitude and influences consumer engagement. The findings provide valuable insights for financial institutions, technology developers, and policymakers.
It presents a clear research model, utilizes appropriate methodologies, and provides practical implications. However, elaborating on the theoretical and practical implications of the findings in relation to existing literature and industry trends would strengthen the paper.
Comments on the Quality of English LanguageOverall, the English in the document is professional and understandable, but there are areas for improvement to enhance clarity and style.
Author Response
Comments 1: This paper explores the impact of generative AI attributes on consumer perceptions and attitudes towards financial advisory services. It employs Service-Dominant Logic and Artificial Intelligence Device Use Acceptance frameworks to analyze data from 822 mobile banking users. The study reveals that personalized investment suggestions, human-like empathy, and continuous improvement of Gen AI positively influence perceived authenticity, which in turn fosters a utilitarian attitude and influences consumer engagement. The findings provide valuable insights for financial institutions, technology developers, and policymakers.
It presents a clear research model, utilizes appropriate methodologies, and provides practical implications. However, elaborating on the theoretical and practical implications of the findings in relation to existing literature and industry trends would strengthen the paper.
Response 1: Response: Thank you for highlighting the need for a more detailed discussion of the theoretical and practical implications of our findings. We agree that a deeper connection to existing literature and industry trends would enhance the contribution of the paper.
In response, we have revised the Discussion and Implications sections of the manuscript to include:
Theoretical Implications: We have expanded our discussion on how the integration of Service-Dominant Logic (SDL) and AI Device Use Acceptance (AIDUA) frameworks provides new insights into the role of generative AI in financial advisory services. Specifically, we explore how our findings build upon and extend existing research in the areas of value co-creation, AI-driven service personalization, and consumer trust in AI. For example, we discuss how the influence of human-like empathy and continuous improvement on perceived authenticity aligns with and contributes to the literature on AI-mediated service interactions (e.g., Pelau et al., 2021). Additionally, we reference Vargo and Lusch’s (2004) SDL framework to explain how AI technologies enable consumers to take an active role in co-creating personalized value.
Practical Implications: In terms of industry trends, we have linked our findings to the growing adoption of AI-driven financial advisory services. We discuss how financial institutions can leverage personalized investment suggestions and human-like empathy to enhance customer trust and engagement, which is particularly important in the context of digital transformation in the finance industry. Moreover, we highlight how the continuous improvement of AI technologies can provide long-term benefits by adapting to evolving consumer needs, which is a key trend in FinTech innovation. We have also added references to industry reports on AI in financial services, which emphasize the practical relevance of our findings for technology developers and policymakers in shaping the future of AI regulation and innovation.
The updated text reads as follows:
6.1 Academic implications
This research significantly enhances the understanding of how Generative AI (GenAI) influences consumer behavior in the realm of financial advice. The study’s findings con-tribute to the theoretical landscape by extending the application of Service-Dominant Log-ic (SDL), integrating the AI Device Use Acceptance (AIDUA) framework, and highlighting the complex interplay between AI attributes and consumer perceptions.
The findings emphasize the importance of personalized investment suggestions, human-like empathy, and continuous improvement in GenAI’s recommendations within the context of consumer value co-creation, as highlighted by SDL theory. By tailoring its services to individual consumer needs and preferences, GenAI facilitates a more interac-tive and collaborative experience between the service provider and the consumer, thus enabling value co-creation. As demonstrated by previous studies (Ref. [90]), personaliza-tion is crucial for enabling value co-creation, allowing for a more interactive and collabo-rative experience between the service provider and the consumer. This study’s findings align with SDL principles and extend the theory by showing how digital technologies en-hance personalized value co-creation, surpassing the limitations of traditional hu-man-to-human service frameworks.
Moreover, GenAI’s ability to exhibit human-like empathy significantly influences consumers’ perceived authenticity by demonstrating genuine care and concern. This finding contributes to the growing body of literature on the importance of designing AI technologies that are not only competent but also genuine and transparent in their inter-actions (Ref. [91]). Additionally, GenAI’s capacity for continuous learning enables it to adapt to evolving user needs and preferences, thereby enhancing its perceived authenticity over time (Refs. [92], [93]).
These findings underscore the importance of integrating personalized investment suggestions, human-like empathy, and continuous improvement in GenAI-driven finan-cial advice. This integration reflects the processes of SDL and AIDUA by co-creating value through tailored, empathetic, and adaptive financial guidance, ultimately enhancing consumer engagement, trust, and participation in GenAI-powered financial services.
The study also highlights the role of perceived authenticity in human-bot interactions, especially within the field of artificial intelligence (Refs. [75], [76]). The positive correlation between GenAI’s features and perceived authenticity aligns with the authenticity princi-ple in AI research (Refs. [29], [94], [95]). This emphasizes the necessity for GenAI and sim-ilar technologies to demonstrate authenticity to effectively engage and support users.
Additionally, the study identifies a strong correlation between perceived authenticity, utilitarian attitudes, and consumers’ willingness or resistance to communicate with GenAI for financial advice. It expands our understanding of technology adoption theories by demonstrating that perceived authenticity enhances utilitarian attitudes towards GenAI, which in turn affects the willingness or resistance to use GenAI for financial ad-vice. This suggests that the value consumers place on authenticity can significantly influ-ence their practical assessment of a technology’s benefits (Ref. [96]). The findings advocate for a broader interpretation of perceived usefulness in AI technology acceptance, high-lighting the importance of authenticity in shaping utilitarian evaluations of AI technology.
Lastly, the study’s focus on AI literacy adds to the theoretical landscape by suggest-ing that a higher level of AI literacy can enhance the effectiveness of AI features in im-proving perceived authenticity and, consequently, utilitarian attitudes (Ref. [97]). This im-plies that individuals’ interactions with AI technologies are significantly influenced by their understanding of the technology, leading to increased acceptance and willingness to communicate with GenAI. Conversely, lower levels of AI literacy may lead to resistance in communicating with GenAI, highlighting the importance of addressing this factor to fa-cilitate the effective integration of AI-driven services in the consumer value co-creation process.
In conclusion, this study offers a comprehensive integration of key concepts, includ-ing personalized investment suggestions, human-like empathy, continuous improvement, perceived authenticity, utilitarian attitudes, and consumers’ willingness or resistance to communicate with GenAI, within the frameworks of SDL and AIDUA. The findings show that GenAI’s personalized and empathetic approach, along with its ability to continuous-ly improve, enhances perceived authenticity and utilitarian attitudes among consumers, facilitating value co-creation as proposed by SDL. Additionally, the study extends the AIDUA model by incorporating continuous improvement as a factor influencing per-ceived authenticity, a key determinant of AI tool usage. The research also underscores the role of AI literacy in shaping consumers’ willingness or resistance to engage with GenAI, highlighting the importance of addressing this factor to ensure the effective integration of AI-driven services in the value co-creation process. Overall, this study contributes to the growing body of literature on AI-driven services and their impact on consumer behavior, providing valuable insights for both researchers and practitioners in the field.
6.2 Practical Implications
The practical implications of this study are substantial, providing valuable insights for a wide range of stakeholders, including financial institutions, technology developers, and policymakers. For financial service providers, the study emphasizes the importance of developing GenAI technologies with enhanced human-like characteristics, such as the ability to offer personalized advice and exhibit empathy. This suggests that financial in-stitutions should invest in AI systems that go beyond basic natural language processing and incorporate the ability to understand and adapt to individual emotional states and preferences. The research indicates that GenAI-driven chatbots capable of recognizing and responding to users' emotions can significantly enhance user satisfaction and engage-ment. This underscores the necessity for financial institutions to employ GenAI technolo-gies that can tailor their services to individual needs and preferences.
Furthermore, the study highlights the importance of continuous learning in main-taining and enhancing consumer trust and engagement with GenAI systems. Financial institutions should prioritize designing AI systems that can continuously update their knowledge base and refine their algorithms based on user interactions. This approach aligns with the continuous improvement aspect of AI development and ensures that AI systems remain relevant and effective in meeting evolving consumer needs and prefer-ences. AI systems capable of continuous learning and improvement are better equipped to build and maintain user trust over time by demonstrating an ongoing commitment to providing accurate and up-to-date information.
The study’s findings also emphasize the importance of AI literacy in enhancing the positive impact of GenAI's attributes on perceived authenticity. This suggests that finan-cial institutions should develop educational programs and resources to improve consum-ers' understanding of AI. By investing in initiatives that demystify AI technologies, finan-cial institutions can reduce resistance and increase engagement among consumers. This aligns with the broader goal of enhancing AI literacy and ensuring that consumers have the necessary knowledge and skills to interact effectively with AI-driven services. Con-sumers with higher levels of AI literacy are more likely to appreciate the benefits of AI-driven services and engage with them more effectively. Therefore, businesses should invest in educational initiatives to promote consumer understanding and acceptance of these technologies.
In conclusion, the study’s implications highlight the importance for policymakers to consider the impact of GenAI-driven financial advice on personalized investment sugges-tions, human-like empathy, and continuous improvement in consumer financial services. As GenAI becomes increasingly integrated into the sector, policymakers must ensure that consumers receive tailored advice that aligns with their unique financial circumstances, fostering trust and engagement. Additionally, they should prioritize consumer privacy protection while promoting equitable access to AI-driven benefits, addressing the digital divide. This may involve establishing standards for transparency in AI algorithms, en-suring data privacy, and implementing digital literacy programs. By proactively address-ing these issues with a focus on personalization, empathy, and continuous improvement, policymakers can create a regulatory landscape that supports responsible innovation. This approach will ultimately encourage the development and deployment of AI technol-ogies within the financial sector that prioritize individual needs, build meaningful con-nections, and continuously evolve to better serve consumers.
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsEnhancing Financial Advisory Services with GenAI: Consumer 2 Perceptions and Attitudes through Service-Dominant Logic and 3 Artificial Intelligence Device Use Acceptance Perspectives
I sincerely appreciate your insightful feedback. All of the review comments have been carefully considered, and the required revisions have been implemented accordingly, accompanied by comprehensive and satisfactory responses.
Reviewer 2 Report
Comments and Suggestions for AuthorsWhile the authors did make some improvements to the paper, the fundamental design of the paper has not changed. As it is, it doesn't show any implications of their findings that a financial institution can utilize. Robot advisors (GAIs are essentially robot advisors) are on the decline in the US due to poor performance and lack of trust by consumers. While the attitude towards GAI in Asian countries are generally more positive than that of the US consumers, is the trust resulted in better returns? There are numerous ways that the study can turn into more rigorous tests of consumer preference and investment returns. The design of the current study is simply too weak.
Let's start with a concrete example: one of the hypotheses is that "Continuous improvement of GAI is positively associate with perceived authenticity." But what is perceived authenticity? It is measured by the three questions in the appendix (PA). Where did those questions come from? I saw references to 24, and 76, two papers that are not closely associated with cognitive psychology, let alone financial management. At best, it is a weak hypothesis based on weak science.
N/A
