A note on how to produce a preliminary dissertation proposal prototype (version 0): for MBA students

A note on how to produce a preliminary dissertation proposal prototype (version 0): for MBA students

The starting point of producing a preliminary dissertation proposal prototype (version 0) is to come up with a crude research theme. A crude research theme is more specific than a management subject name (e.g. supply chain management, human resource management). To get some idea of a crude research theme, an MBA student could ask a Generative AI tool (including asking Google is also OK) to learn what are the main topics of a management subject. For example:

I ask Google: "Suggest 6 main topics of Human Resource Management".

The reply from Google is: "Human Resource Management (HRM) encompasses the strategic practices used to manage a company’s workforce. Its six core topics include talent acquisition, performance management, learning and development, compensation and benefits, employee relations, and compliance...."

Now, the MBA student feels like choosing the topic of learning and development to work with; the student then could contextualize the topic a bit more to become a crude research theme, such as "reviewing the learning and development policy of the Hong Kong Banking sector in response to the impacts of AI". Next, the student needs to express some additional research preference in the form of a few key words. To do so, the student could ask Google to suggest 5 subtopics of the crude research theme. This can be done like this:

I ask Google: "Suggest 5 subtopics of the crude MBA dissertation research theme of "reviewing the learning and development policy of the Hong Kong Banking sector in response to the impacts of AI".

The reply from Google is: "These 5 subtopics break down the broad dissertation theme into manageable, academically rigorous areas, addressing both macro-level strategy and micro-level practitioner impacts in the Hong Kong banking sector.

1. Regulatory Alignment and Compliance Training

This subtopic investigates how banks align their internal Learning and Development (L&D) frameworks with the Hong Kong Monetary Authority’s (HKMA) Manpower Management in the Age of Artificial Intelligence guidelines and the Policy Statement on Responsible Application of Artificial Intelligence. It focuses on upskilling staff to meet strict ethical and risk-management standards.

2. The Generative AI (GenAI) Skills Gap

Research can focus on the disparity between existing employee skill sets and the demands of widespread GenAI adoption. This area evaluates L&D strategies used by local institutions to close the talent gap Hong Kong banks race to hire AI-ready talent as skill gap widens in core areas like credit evaluation, algorithmic accountability, and financial-crime compliance.

3. Assessing the ROI of AI Training Programs

This section measures the effectiveness and return on investment of various AI training initiatives Hong Kong banks race to hire AI-ready talent as skill gap widens. It explores whether current upskilling efforts—ranging from external programs like the HKIB Latest CPD Training Courses to in-house vendor training—are tangibly improving employee productivity and service quality.

4. Change Management and Corporate Culture Transformation

This subtopic explores how banks manage the cultural shift required for an "AI-human" collaborative workforce. It focuses on L&D approaches designed to mitigate staff resistance, alleviate fears of job displacement, and promote a "test and learn" mindset supported by local initiatives like the Transformation of Hong Kong's Banking Sector under “Fintech 2025” GenAI Sandbox.

5. Future-Proofing Career Pathways

Instead of focusing solely on technical skills, this area analyzes how L&D policies are adapting to redefine traditional banking roles. It examines training curriculums for "human-centric" competencies such as emotional intelligence, complex problem-solving, and relationship management that AI cannot easily replicate."

On reviewing the reply from Google, I now made up my mind on my crude research theme, which is ""reviewing the learning and development policy of the Hong Kong Banking sector in response to the impacts of AI". I also adopt the two key words of The Generative AI (GenAI) Skills Gap, Change Management and Corporate Culture Transformation. I also include the key word of pragmatism because the MBA program is mainly pragmatic in orientation. 

Now, the student is ready to go to a generative AI platform and tell it his/her query:

The student query: "I am a part-time MBA student learning to formulate my dissertation proposal. My initial crude research theme is: ""reviewing the learning and development policy of the Hong Kong Banking sector in response to the impacts of AI". I am also interested in considering the following key words to refine my research theme: The Generative AI (GenAI) Skills Gap, Change Management and Corporate Culture Transformation, and pragmatism. Please suggest some relevant research issues, some Hong Kong public media-reported issues, and 4 relevant academic theories with Harvard references on them. Also suggest 3 more concrete research themes based on the crude research theme and the three key words".

I posted the query to a generative AI and got the following reply: "Your topic is promising, but it needs to be narrowed from a broad policy review into a focused dissertation question. A strong direction is to study how Hong Kong banks are reskilling staff and reshaping learning-and-development policy to cope with GenAI adoption, talent gaps, and culture change in a regulated environment.

Relevant research issues

The most relevant research issues are not just “whether AI is used,” but how banks are preparing people and organisations for it. Hong Kong evidence shows GenAI adoption is already progressing, but banks still face constraints around model accuracy, privacy/security, and talent/resources. The HKMA has also explicitly signalled that banks should adjust training, reskilling, and staff support in light of technological change.

Useful research issues include:

• How Hong Kong banks are redesigning learning and development policies for AI-enabled work.

• Whether current training is closing the GenAI skills gap or only providing basic awareness.

• How banks are balancing upskilling with governance, compliance, and operational risk.

• What cultural barriers affect employee acceptance of AI-enabled working practices.

• How senior management and boards support AI-related capability building.

• Whether training programmes are aligned with actual role changes, not just technology adoption.

Hong Kong issues from media

Several Hong Kong public media and public-sector reports point to concrete dissertation-worthy issues. The Hong Kong banking sector is being pushed toward AI adoption while regulators stress responsible use, governance, and human oversight. A 2025 HKIMR/AoF report said 75% of surveyed financial institutions had already implemented or were piloting GenAI use cases, with adoption expected to rise to 87% in three to five years.

Publicly reported issues you can use:

• A widening AI and talent shortage in banking, especially around AI, green finance, and new-market skills.

• The need for stronger internal training rather than relying only on hiring from competitors.

• The HKMA’s push for capacity-building and reskilling, including guidance on manpower development and training.

• Governance and trust concerns, including data privacy, security, explainability, and model validation.

• The challenge of building an AI-ready culture, not just deploying tools; one Hong Kong report noted that wider deployment is held back by talent shortages and organisational rigidity.

• Growing official support for responsible AI adoption through policy statements, sandboxing, and supervisory guidance.

Four academic theories

Here are four theories that fit your topic well, with Harvard-style references.

1. Dynamic capabilities theory

   This theory helps explain how banks sense technological change, seize opportunities, and reconfigure capabilities through training and organisational change. It is especially useful for studying why L&D policy must evolve when AI changes work tasks.
   Reference: Teece, D.J., Pisano, G. and Shuen, A. (1997) ‘Dynamic capabilities and strategic management’, Strategic Management Journal, 18(7), pp. 509–533.

2. Absorptive capacity theory

   This theory is useful for explaining how banks identify, assimilate, and apply AI-related knowledge through training and learning systems. It fits your focus on the GenAI skills gap and on whether banks can actually turn external AI knowledge into internal competence.
   Reference: Cohen, W.M. and Levinthal, D.A. (1990) ‘Absorptive capacity: a new perspective on learning and innovation’, Administrative Science Quarterly, 35(1), pp. 128–152.

3. Organisational change theory / change management

   This theory helps analyse how banks manage resistance, redesign roles, and communicate change during AI adoption. It is directly relevant to culture transformation and to whether L&D policy is part of a broader change programme.
   Reference: Kotter, J.P. (1996) Leading Change. Boston: Harvard Business School Press.

4. Institutional theory

   This theory is useful because Hong Kong banks operate in a heavily regulated environment where policy, professional norms, and supervisory expectations strongly shape behaviour. It can help you study how HKMA guidance, industry norms, and legitimacy pressures influence L&D policy choices.
   Reference: DiMaggio, P.J. and Powell, W.W. (1983) ‘The iron cage revisited: institutional isomorphism and collective rationality in organizational fields’, American Sociological Review, 48(2), pp. 147–160.

Three more concrete themes

Based on your crude theme and keywords, these would be stronger and more researchable:

1. A pragmatist study of L&D policy responses to the GenAI skills gap in Hong Kong retail banks.

   This keeps your policy focus but narrows it to a clear problem: how banks redesign training to match GenAI capability needs.

2. How Hong Kong banks use change management to build an AI-ready learning culture under HKMA guidance.

   This lets you examine organisational culture transformation, staff resistance, leadership support, and the role of supervisory expectations.

3. Evaluating the effectiveness of reskilling and upskilling strategies for GenAI adoption in Hong Kong banks: a pragmatic case study.

   This is methodologically strong for an MBA dissertation because you can combine interviews, document analysis, and perhaps a small survey under a pragmatist approach.

Best refinement

A strong dissertation title would be something like: “How Hong Kong retail banks are adapting learning and development policy to address GenAI-related skills gaps: a pragmatist case study of organisational change and culture transformation.” This is specific, current, and well supported by public-sector and media evidence on AI adoption, skill shortages, and reskilling priorities."

After further refining the research theme, the student could post the refined research theme, and the key words to a generative AI platform, and asks it produce a preliminary proposal prototype (preferably also mention your basic relevant profile [the researcher positionality]) with the following elements: dissertation project title, 2 related research issues and 2 related public media-reported issues, 4 research objectives, 4 research questions, 2 primary research methods, 1 secondary research methods, 2 academic theories to use (with references in Harvard reference format). [bear in mind that these information elements are what you fill in the synopsis form].

** a related blog note: The rationale of doing an exploratory study before brainstorming with poe to produce a proposal prototype; another related note is the spiral model to produce a proposal prototype, and subsequently the polished proposal.

** also study a note on converting a concrete research theme into a full-fledge dissertation proposal prototype: an example.

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