Sunday, 31 May 2026

An example of an exploratory attempt to brainstorm on an MBA research theme about consumer trust on AI-enabled financial services in Hong Kong

An example of an exploratory attempt to brainstorm on an MBA research theme about consumer trust on AI-enabled financial services in Hong Kong

 

I am a part-time MBA student currently learning to do my dissertation project. My research theme is "an assessment of consumer trust of gen x and gen z (notably on transparency, data privacy and fairness) in innovative AI-enabled financial services in Hong Kong from the pragmatic perspective" Could you describe the main research issues, the main public media-reported issues and 3 main academic theories as related to this research theme.

Your topic is well framed for a Hong Kong MBA dissertation: the core issues are trust formation and trust erosion around AI-enabled financial services, with age-group differences likely to matter most on transparency, privacy, fairness, and perceived control. In Hong Kong, public and regulatory discussion is already centered on model accuracy, data privacy/security, transparency, and bias/fairness in AI finance.

Main research issues

1.    What “trust” means in AI finance.
You will need to define consumer trust clearly: trust in the institution, trust in the AI system, or trust in the outcome of the service. In your topic, trust likely includes whether users believe the AI is accurate, honest, explainable, secure, and fair.

2.    Generational differences.
Gen Z and Gen X may not trust the same cues. Hong Kong evidence suggests Gen Z tends to demand novelty, product fit, and clear explanations, while Gen X leans more toward reliability and proven brand reputation. This makes generational comparison a central research issue rather than a simple control variable.

3.    Transparency and disclosure.
A major question is whether users want to know when AI is being used, how decisions are made, and what data is collected. Hong Kong policy discussion emphasizes disclosure and transparency when AI affects customers or investors, and recent consumer reporting suggests Hong Kong users want explicit identification of AI use.

4.    Data privacy and security.
This is likely one of the strongest trust drivers in financial services because the context involves sensitive personal and financial data. Hong Kong’s applied research and policy statements repeatedly identify privacy, security, and compliance as adoption barriers and governance priorities.

5.    Fairness and bias.
Consumers may worry that AI systems treat people unevenly, discriminate indirectly, or make opaque decisions. Hong Kong AI governance materials explicitly frame fairness as avoiding favoritism, discrimination, and harm, so fairness is not just ethical but also a practical trust issue.

6.    Human oversight versus automation.
Another issue is whether consumers want AI alone, or AI with human review. Hong Kong consumer reporting shows many people still prefer access to a human agent when needed, which suggests that “human-in-the-loop” design may support trust.

Main media-reported issues

·        Synthetic deception and identity risk.
Media coverage in Hong Kong has highlighted that many consumers cannot tell AI-generated voices from human staff, raising concerns about impersonation, fraud, and reputational damage.

·        Low tolerance when AI underperforms.
Reports suggest Hong Kong consumers are willing to use AI, but they react badly when service quality falls short, especially in customer service settings.

·        Demand for clear AI disclosure.
Regional consumer reporting indicates that people want brands to disclose AI usage, and they see this as important for trust. That issue is especially relevant in finance, where hidden automation can feel misleading.

·        Privacy and data use concerns.
Public discussion in Hong Kong repeatedly returns to data privacy, security, and the use of personal data for model training or personalization. This is especially sensitive in banking, insurance, and payments.

·        Fairness, bias, and regulation.
Hong Kong’s AI governance discourse emphasizes fairness, transparency obligations, and risk-tiered controls, showing that public debate is not only about innovation but also about preventing discriminatory or harmful outcomes.

Three useful theories

1.    Trust Theory / Institutional Trust.
This helps explain why consumers rely on financial institutions and AI systems under uncertainty. It is especially useful for examining perceived competence, integrity, and benevolence, which map neatly onto your themes of transparency, privacy, and fairness.

2.    Technology Acceptance Model or UTAUT.
These frameworks explain adoption through perceived usefulness, ease of use, social influence, and facilitating conditions. They are useful because AI financial services are only trusted if people think they are useful, easy, and safe enough to use; they also help explain why Gen Z and Gen X may adopt for different reasons.

3.    Organizational Justice / Fairness Theory.
This theory is highly relevant because users evaluate whether decisions and procedures are fair, consistent, and explainable. In AI finance, it helps you study whether consumers view automated decisions as procedurally fair and whether they feel treated without bias.

Pragmatic perspective

A pragmatic perspective fits your dissertation well because it lets you focus on what works in practice rather than forcing one philosophical position. In your case, pragmatism supports a mixed understanding of trust: you can study both measurable patterns, such as survey differences between Gen X and Gen Z, and deeper explanations, such as why transparency or privacy concerns matter in context. That makes your theme suitable for a research design that links real-world problems, consumer experience, and actionable recommendations for Hong Kong financial firms.

A workable framing

A strong dissertation framing could be: “How do transparency, data privacy, and fairness shape Gen X and Gen Z consumer trust in AI-enabled financial services in Hong Kong, and what practical governance features increase trust?” That framing stays close to public policy concerns in Hong Kong while giving you a clear analytical structure.

 

For the academic theories suggested, pleas provide a few references on them in Harvard reference format.

Here are key references for the three main theories, in Harvard style, with a mix of classic/theory-defining sources and more recent applications to AI, finance, or privacy:


1. Trust Theory (organizational/consumer trust)

Mayer, R.C., Davis, J.H. and Schoorman, F.D. (1995) An integrative model of organizational trust. Academy of Management Review, 20(3), pp. 709–734. https://doi.org/10.5465/amr.1995.9508080335

Gulati, R. and Singh, H. (2024) Advice by algorithms: dual dimensions of trust in AI financial service adoption. Journal of Retailing and Consumer Services, 88, p. 103169. https://doi.org/10.1016/j.jretconser.2024.103169

Farrell, A., et al. (2016) Trust in artificial intelligence: a comprehensive conceptual review. Electronic Markets, 32(4), pp. 1–22. https://doi.org/10.1007/s12525-022-00592-6


2. Technology Acceptance Model (TAM) / UTAUT

Davis, F.D. (1989) Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13(3), pp. 319–340. https://doi.org/10.2307/249008

Venkatesh, V., Morris, M.G., Davis, G.B. and Davis, F.D. (2003) User acceptance of information technology: toward a unified view. MIS Quarterly, 27(3), pp. 425–476. https://doi.org/10.2307/30036540

The impact of AI on UX in fintech payment systems for Gen Z. (2025) Theseus, Helsinki. Available at: https://www.theseus.fi/handle/10024/880931

Digital banking adoption in emerging economies: revisiting TAM. (2025) Quarterly Journal of Social Sciences, pp. 1–20. Available at: https://qjss.com.pk/index.php/qjss/article/view/366


3. Organizational Justice / Fairness Theory (including procedural fairness in privacy)

Colquitt, J.A. (2001) On the dimensionality of organizational justice: a construct validation of a measure. Journal of Applied Psychology, 86(3), pp. 386–400. https://doi.org/10.1037/0021-9010.86.3.386

Culnan, M.J. and Bies, R.J. (2003) Consumer privacy: balancing economic and justice considerations. Journal of Social Issues, 59(2), pp. 323–342. https://doi.org/10.1111/1540-4560.00070

Culnan, M.J. and Armstrong, P.K. (1999) Information privacy concerns, procedural fairness, and impersonal trust: an empirical investigation. Organization Science, 10(1), pp. 104–115. https://doi.org/10.1287/orsc.10.1.104

Hong Kong’s policy statement on responsible application of AI in the financial market. (2024) Financial Services and the Treasury Bureau, Hong Kong SAR. Available at: https://gia.info.gov.hk/general/202410/28/P2024102800154_475819_1_1730083937115.pdf


These references give you:

·        The foundational trust model (Mayer et al., 1995) and recent AI/finance trust work.

·        The original TAM (Davis, 1989), UTAUT (Venkatesh et al., 2003), and recent applications to fintech/Gen Z.

·        Core organizational justice/fairness work plus privacy-specific fairness and trust (Culnan & Bies; Culnan & Armstrong), and Hong Kong AI policy context for your setting.

 

As my additional preference is on the research issues of (1) data privacy and security and (2) human oversight vs automation) and that my favored primary research methods to use are (1) focus group with 6 participants, two being Gen X financial services consumers and two being Gen Z financial services consumers, one being an IT manager with a financial service firm in Hong Kong and one being a university lecturer on IT AND (2) an opinion questionnaire survey to my friends. Also bear in mind that this is a part-time MBA dissertation project. Could you suggest a dissertation proposal to me, notably with 4 research objectives and 4 research questions.

Below is a focused, MBA‑appropriate dissertation proposal tailored to your preferences: two priority research issues (data privacy & security; human oversight vs automation), a pragmatic mixed‑methods design, one small focus group (6 participants: 2 Gen X consumers, 2 Gen Z consumers, 1 IT manager from a Hong Kong financial firm, 1 IT lecturer), and an opinion questionnaire survey to your friends. It includes four research objectives and four research questions.


Dissertation proposal (part‑time MBA)

Working title
An assessment of consumer trust of Gen X and Gen Z in innovative AI‑enabled financial services in Hong Kong from a pragmatic perspective: the roles of data privacy & security and human oversight versus automation

Research theme and context
AI‑enabled financial services in Hong Kong are expanding rapidly, but adoption is constrained by concerns over data privacy, security, transparency, and fairness. Consumers' trust is critical for adoption, yet Gen X and Gen Z may differ in what drives trust. Hong Kong's AI policy and industry reports emphasize privacy/security, fairness, and transparency as core governance principles.


Research problem

Despite growth in AI applications in finance, there is limited empirical evidence on how Gen X and Gen Z consumers in Hong Kong trust AI‑enabled financial services, especially regarding:

·        Data privacy and security (how data is collected, used, protected), and

·        Human oversight vs automation (degree of human review, ability to reach a human agent, accountability).

This gap is important for managers designing AI services and governance practices that build trust across generations.


Research philosophy

Pragmatism: A pragmatic worldview is well‑suited to mixed‑methods research and applied managerial problems. It focuses on “what works” to solve real‑world problems, allowing you to combine qualitative insights (focus group) with quantitative patterns (survey) to inform practical recommendations.


Theoretical framework

You will draw on:

1.    Mayer et al.'s integrative model of organizational trust (competence, integrity, benevolence) to frame trust in AI/financial institutions.

2.    Technology Acceptance Model (TAM) / UTAUT (perceived usefulness, ease of use, trust, security, privacy risk) to explain adoption and trust in AI financial services.

3.    Organizational justice / fairness theory (procedural, distributive, interactional justice) to examine fairness, transparency, and accountability in AI decisions.

These theories align with your focus on transparency, privacy, security, fairness, and human oversight.


Research objectives

1.    RO1: To examine how Gen X and Gen Z consumers in Hong Kong perceive data privacy and security risks in AI‑enabled financial services and how these perceptions affect their trust.

2.    RO2: To explore how preferences for human oversight versus full automation shape consumer trust in AI‑enabled financial services for Gen X and Gen Z.

3.    RO3: To compare differences between Gen X and Gen Z in the relative importance of privacy/security, human oversight, transparency, and fairness for trust in AI financial services.

4.    RO4: To develop practical recommendations for financial service firms in Hong Kong on designing AI services and governance (e.g., disclosure, human‑in‑the‑loop, security measures) that enhance consumer trust across generations.


Research questions

1.    RQ1: How do Gen X and Gen Z consumers in Hong Kong perceive data privacy and security risks in AI‑enabled financial services, and how do these perceptions influence their trust?

2.    RQ2: How do preferences for human oversight versus automation affect consumer trust in AI‑enabled financial services for Gen X and Gen Z?

3.    RQ3: In what ways do Gen X and Gen Z differ in the relative importance of privacy/security, human oversight, transparency, and fairness for trust in AI financial services?

4.    RQ4: What practical governance and design practices can Hong Kong financial service firms adopt to strengthen consumer trust in AI‑enabled services across Gen X and Gen Z?


Methodology (pragmatic mixed‑methods)

Overall design

·        Qualitative strand: One semi‑structured focus group (6 participants) to explore in depth how different stakeholders understand trust, privacy/security, and human oversight in AI finance, and to generate rich insights and hypotheses.

·        Quantitative strand: An opinion questionnaire survey distributed to your friends to measure attitudes, perceived risks, and trust levels related to AI financial services, and to compare Gen X vs Gen Z at a basic level.

·        Integration: Use focus group findings to inform survey items and interpretation; use survey results to check whether qualitative themes appear more broadly. This triangulation supports a pragmatic, problem‑centered approach.

Participants and sampling

Focus group (n = 6)

·        2 Gen X consumers of financial services (born ~1965–1980), Hong Kong residents

·        2 Gen Z consumers of financial services (born ~1997–2012), Hong Kong residents

·        1 IT manager from a Hong Kong financial service firm

·        1 university lecturer specializing in IT (possibly with interest in AI/finance)

Sampling: purposive, based on generation, role, and relevance to AI/finance.

Survey

·        Convenience sample of your friends (likely mixed generations, including Gen X and Gen Z).

·        You will record basic demographics (generation, age, gender, use of AI financial services, experience level).

·        You will acknowledge limitations (non‑probability sample, potential bias) and frame findings as exploratory, not generalizable to the whole population.

Data collection instruments

Focus group guide (semi‑structured, ~90 minutes)
Key topics:

·        Experiences with AI‑enabled financial services (chatbots, robo‑advisors, AI‑driven credit decisions, fraud detection).

·        Perceptions of data privacy and security:

·        What data is collected?

·        How is it used?

·        Concerns about breaches, misuse, surveillance.

·        Human oversight vs automation:

·        Preferences for human review of AI decisions.

·        Expectations of access to a human agent when things go wrong.

·        Views on accountability and explainability.

·        Transparency and fairness:

·        What information do they want about AI use?

·        What feels fair or unfair in automated decisions?

·        Generational differences:

·        How do Gen X and Gen Z views differ?

·        What builds or erodes trust for each group?

Survey questionnaire (online, ~15–20 minutes)

Sections:

1.    Demographics: generation, age, gender, occupation, financial service usage.

2.    AI financial service experience: types used, frequency, confidence.

3.    Data privacy & security perceptions (Likert scale items, e.g., 1–5):

·        “I am concerned that AI financial services collect too much personal data.”

·        “I worry about data breaches in AI‑enabled financial services.”

·        “I trust Hong Kong financial firms to protect my data in AI systems.”

4.    Human oversight vs automation:

·        “I prefer AI decisions in finance to be reviewed by a human before being final.”

·        “I want the option to speak to a human when using AI financial services.”

·        “Fully automated financial advice gives me less trust than human advisors.”

5.    Transparency & fairness:

·        “I want clear disclosure when AI is used in financial decisions.”

·        “I am concerned that AI may treat people unfairly.”

6.    Trust in AI financial services (outcome variable):

·        “I trust AI‑enabled financial services in Hong Kong.”

·        “I would recommend AI financial services to friends/family.”

7.    Generational comparison items: optional open‑ended question on what would make them trust AI finance more.

You will pilot the survey with 2–3 people to check clarity.

Data analysis

Focus group

·        Transcribe (or use detailed notes) and code responses thematically.

·        Identify themes around:

·        Privacy & security concerns

·        Human oversight preferences

·        Transparency and fairness

·        Generational differences

·        Use simple matrices to compare Gen X vs Gen Z views.

Survey

·        Basic descriptive statistics (means, frequencies) for all items.

·        Compare Gen X vs Gen Z on key constructs using simple group comparisons (e.g., mean differences).

·        Optionally, simple correlation or regression to see how privacy/security concerns and human oversight preferences relate to trust.

·        Present results in tables and brief narratives.

Integration

·        Use focus group insights to explain survey patterns (e.g., why Gen Z rates privacy concerns higher/lower).

·        Highlight convergences and divergences between qualitative and quantitative findings.


Expected contributions

Managerial/practical contributions

·        Clear guidance for Hong Kong financial firms on:

·        How to design privacy‑by‑default and security measures to build trust.

·        How to balance automation with human oversight and human‑agent access.

·        What transparency and fairness features matter most to Gen X and Gen Z.

·        Practical recommendations for AI governance aligned with Hong Kong's responsible AI principles (fairness, privacy & security, transparency).

Academic contributions

·        Empirical evidence on generational differences in trust in AI finance in Hong Kong.

·        Application of trust theory, TAM/UTAUT, and organizational justice to a new context (AI‑enabled financial services).

·        A pragmatic mixed‑methods approach that can be extended in future research with larger samples.


Ethical considerations

·        Obtain informed consent from all focus group participants and survey respondents.

·        Ensure anonymity and confidentiality (no names in analysis; use pseudonyms for focus group quotes).

·        State that participation is voluntary and that respondents can withdraw at any time.

·        For the survey to friends, explicitly state that:

·        It is for academic research.

·        Data will be used only for the dissertation.

·        Results will be reported in aggregate.

·        Obtain any required ethics approval from your MBA program before data collection.


Limitations

·        Small focus group: 6 participants limits depth and generalizability; findings are exploratory and illustrative rather than definitive.

·        Convenience survey: Friends are not a random sample; results may be biased and not representative of all Hong Kong consumers. Frame findings as preliminary and use them to generate insights, not strong general claims.

·        Self‑report data: Attitudes may not match actual behavior.

·        Single city context: Results apply to Hong Kong and may not generalize to other regions.

You will address these by:

·        Being transparent about limitations.

·        Using triangulation (focus group + survey) to strengthen confidence in key themes.

·        Positioning the study as an applied, managerial investigation rather than a large‑scale epidemiological study.


Proposed chapter structure

1.    Introduction

·        Background, problem statement, research objectives and questions, significance, scope.

2.    Literature review

·        Consumer trust in AI; generational differences (Gen X vs Gen Z); privacy/security; human oversight vs automation; transparency and fairness.

·        Theoretical framework: trust theory, TAM/UTAUT, organizational justice.

·        Hong Kong context: AI in finance, regulatory and policy environment.

3.    Methodology

·        Research philosophy (pragmatism), design, participants, instruments, procedures, data analysis, ethics, limitations.

4.    Findings

·        Focus group results (themes, quotes, generational comparisons).

·        Survey results (descriptive statistics, Gen X vs Gen Z comparisons).

·        Integrated discussion of findings.

5.    Discussion

·        Interpretation in relation to theory and literature.

·        Managerial implications for Hong Kong financial firms.

·        Recommendations for AI governance and service design.

6.    Conclusion

·        Summary of key findings.

·        Contributions.

·        Limitations and suggestions for future research.

 



A collection of blog notes on using chatgpt for research purpose.


No comments:

Post a Comment