INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025
Page 743
www.rsisinternational.org
Enhancing Competency and Performance in Wealth Management
Education through Artificial Intelligence (AI) powered Virtual
Reality (VR) Learning
Muhammad Shodiq
PhD Student, International Islamic University Malaysia
DOI: https://doi.org/10.47772/IJRISS.2025.91200063
Received: 09 December 2025; Accepted: 17 December 2025; Published: 31 December 2025
ABSTRACT
The rapid evolution of financial markets demands a corresponding advancement in wealth management
education, particularly in developing competencies that align with real-world performance expectations. This
study explores the integration of Artificial Intelligence (AI) and Virtual Reality (VR) technologies to enhance
learning outcomes in wealth management training especially on the Market Linked Deposit Product. By
leveraging AI-driven personalization and immersive VR simulations, learners engage in realistic financial
advisory scenarios that foster critical thinking, decision-making, and client interaction skills. A mixed-methods
approach was employed, combining quantitative performance assessments with qualitative feedback from
participants for the Relationship Manager at CIMB Niaga. Results indicate significant improvements in learner
engagement, competency acquisition, and practical application of wealth management principles as well as the
improvements of business outcome. The findings suggest that AI-powered VR learning environments offer a
scalable and effective solution for bridging the gap between theoretical knowledge and professional readiness
in financial education. Implications for curriculum design, instructional strategies, and future research
directions are discussed.
Keywords: AI in education, Virtual Reality, Wealth Management, Competency Development, Performance
Enhancement
INTRODUCTION
Indonesia's wealth management industry is poised for significant growth, driven by a rising middle class,
increasing GDP per capita, and a growing appetite for diversified financial products. However, the sector faces
persistent challenges in education, talent development, and product comprehension—particularly regarding
complex instruments like Market Linked Deposits (MLDs). Despite Indonesia's promising economic
indicators, the AUM (Assets Under Management) penetration remains low at around 4% of GDP, compared
to 15–25% in neighboring countries such as Malaysia and Thailand. This gap reflects both limited investor
engagement and a shortage of qualified financial advisors capable of delivering holistic wealth management
services (Smallwood, M., Prosowicz, G., Dirga, A., et al. (2024). MLDs are structured products that combine
traditional deposit features with derivative-linked returns. While they offer principal protection and potential
upside, they are not covered by Indonesia’s Deposit Insurance Corporation (LPS), making them riskier and
harder to explain to retail clients. The lack of standardized training and certification around these products
contributes to mis-selling risks and regulatory scrutiny (Gunardi, H., Primiana, I., Effendi, N., Herwany, A.
(2024). A study involving over 1,800 Indonesian students revealed that financial literacy remains critically
low, with 79.4% unaware of national financial education tools. This highlights the urgent need for integrated
financial education programs that go beyond basic budgeting and saving to include investment literacy and risk
management (Senduk, W.F., Djatmika, E.T., Wahyono, H., et al. (2024).
The Financial Services Authority (OJK) has introduced tighter regulations to improve transparency and
consumer protection, especially in the sale of structured products. These changes require advisors to be more
competent in product suitability analysis, risk profiling, and ethical advisory practices.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025
Page 744
www.rsisinternational.org
The banking industry is evolving rapidly due to technological advancements, regulatory changes, and shifting
customer expectations. To stay competitive, financial institutions must ensure that their workforce is equipped
with the latest knowledge and skills. Implementing an effective learning strategy is crucial to drive innovation,
improve employee performance and productivity, enhance customer experience, strong risk management and
maintain compliance with Bank regulations, which are will contribute to revenue growth and sustainable
growth.
Traditional banking training methods in Indonesia, such as classroom-based lectures and on-the-job training,
often lack interactivity and adaptability, limiting their effectiveness in addressing diverse learning needs. With
the increasing complexity of banking operations and the demand for digital literacy, there is a need for
innovative tools to bridge the gap between theoretical knowledge and practical application. Learning strategies
refer to structured approaches that employees and organizations use to acquire, process, and apply knowledge
effectively. These strategies, such us personalized learning, continuous upskilling, experiential learning, that
help organization to stay competitive in the era of digital transformation and industry change rapidly.
AI- Powered VR Learning
VR and AI have shown promise in educational settings, their application in banking education in Indonesia
remains underexplored. There is a lack of empirical evidence on how these technologies impact competency
development and performance outcomes in this context. One of the most significant trends in banking learning
and development is the implementation of artificial intelligence (AI) and machine learning. These technologies
are being used to analyze customer data, improve the customer experience, and enhance risk management
processes. By utilizing AI and machine learning, banks can provide personalized learning experiences for their
employees and better understand customer behaviors and preferences. Using the AI based-technology to create
personalized learning paths for employees based on their roles, skill levels, and career aspirations.
Personalized learning paths can help employees feel more engaged and invested in their development.
Digital transformation, including AI-powered Virtual Reality (VR) learning platforms, offers a scalable
solution to address competency gaps. These technologies can simulate real-world advisory scenarios, enhance
product understanding, and personalize learning experiences for both novice and experienced advisors. Such
innovations are essential to prepare professionals for the evolving demands of wealth management in
Indonesia.
The integration of Artificial Intelligence (AI) and Virtual Reality (VR) into education is reshaping traditional
pedagogical models, offering immersive, personalized, and data-driven learning experiences that were
previously unattainable.
AI technologies—such as machine learning, natural language processing, and intelligent tutoring systems—are
increasingly used to support adaptive learning, automate assessments, and personalize instruction. AI enables
real-time feedback, predictive analytics for student performance, and dynamic content delivery tailored to
individual learning styles. According to Ifenthaler et al. (2024), AI's transformative potential lies in its ability
to augment human capabilities rather than replace them, emphasizing the importance of "humans in the loop"
for ethical and effective educational outcomes.springer
The ISAR model proposed by Bauer et al. (2025) categorizes AI's impact into four
dimensions: Inversion, Substitution, Augmentation, and Redefinition. AI can substitute traditional methods,
augment cognitive support, or redefine learning tasks to foster deeper engagement. However, successful
implementation depends on educators' technological literacy and students' readiness to interact meaningfully
with AI systems.
Immersive Virtual Reality (I-VR) offers high-fidelity simulations that allow learners to engage with complex
environments and scenarios in a safe, controlled setting. VR has proven especially effective in procedural
learning, such as medical training, engineering, and financial simulations. A systematic review by Hamilton et
al. (2021) found that most studies reported significant improvements in learning outcomes when VR was used
compared to traditional methods, particularly in science and technical education.springer
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025
Page 745
www.rsisinternational.org
VR's key strengths include:
Immersion: Learners experience a sense of presence that enhances engagement and retention.
Interactivity: Users can manipulate virtual environments, test hypotheses, and receive immediate feedback.
Accessibility: VR can simulate environments that are logistically or financially inaccessible in real life.
When combined, AI and VR create intelligent immersive environments that adapt to learners' behaviors and
decisions. For example, in wealth management education, AI can analyze learner responses during a VR-based
client advisory simulation and adjust the scenario complexity in real time. This synergy supports experiential
learning, critical thinking, and decision-making—key competencies in financial advisory roles.
In the context of Indonesian banking practices, where products like Market Linked Deposits (MLDs) require
nuanced understanding and ethical advisory, AI-powered VR can simulate client interactions, risk profiling,
and product suitability analysis. This approach not only enhances technical knowledge but also builds soft
skills such as communication and empathy, which are essential for trusted financial advisory.
Research Gap and Rationale for Integrating AI-Powered VR in Wealth Management Education
Despite the growing body of research on AI and VR in education, several gaps remain. First, many studies
focus on either AI or VR independently, with limited empirical investigation into their integrated impact on
competency and performance outcomes. Second, there is a lack of domain-specific research examining AI-
powered VR in professional education contexts, particularly within complex decision-making disciplines.
Furthermore, existing studies often emphasize technological feasibility rather than pedagogical effectiveness
and learning analytics integration. This highlights the need for research that examines how AI-powered VR
learning environments can be systematically designed, implemented, and evaluated to enhance learner
competency and performance.
Despite the increasing complexity of financial products such as Market Linked Deposits (MLDs) and the
growing demand for personalized financial advisory services, current wealth management education in
Indonesia remains largely traditional, relying on classroom-based instruction and static content delivery. This
approach fails to adequately prepare learners for real-world advisory scenarios that require dynamic decision-
making, ethical judgment, and client-centric communication.
Existing studies have explored the use of AI and VR in general education and technical training, but there is a
notable gap in applying these technologies to financial education, particularly in emerging markets like
Indonesia. Few programs offer immersive, scenario-based learning that simulates client interactions, product
suitability analysis, and regulatory compliance—all critical competencies for wealth advisors.
Moreover, while AI has shown promise in adaptive learning and VR in experiential training, their combined
potential remains underexplored in the context of Market Linked Deposit. This paper addresses this gap by
investigating how AI-powered VR learning environments can enhance competency and performance in wealth
management education, with a focus on MLD advisory practices.
Research Questions
RQ1: How does the integration of AI-powered VR learning environments affect competency development in
wealth management education?
RQ2: What is the impact of immersive learning on learners’ understanding and advisory performance related to
Market Linked Deposit products?
RQ3: How do learners perceive the effectiveness and engagement of AI-VR-based training compared to
traditional methods?
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025
Page 746
www.rsisinternational.org
Hypotheses
H1: Learners exposed to AI-powered VR simulations will demonstrate significantly higher competency in
wealth management advisory tasks than compared before implementing AI -powered VR learning through
conventional methods.
H2: AI-VR learning environments will improve learners’ understanding of complex financial products, such as
MLDs, more effectively than static instructional materials.
H3: Participants will report higher engagement and satisfaction with AI-VR training compared to traditional
classroom-based education.
LITERATURE REVIEW
Computer-based technology in general has the potential to greatly impact the ways in which we educate and
train all people. Furthermore, computer-based technologies have the great potential to customize and tailor
training material to every individual need. This not only includes the formidable advantages of computer-based
tutors that allow an individual a great deal of control over the pace and style of learning, but also the chance to
access materials when and where they require the training (now called "just in time" training). Hence, it is now
possible to plan strategies for continuous learning at home and on the job, with whatever depth an individual
desires.
Computer-based technologies offer us the means of extending and enhancing every costly hour spent with live
banking operational environments in banking field exercises. Rather than reducing the value of the necessary
field experiences, VR and other computer technologies can insure that the value of these hours are enhanced
by both providing the much needed training prior to the field hours, and then again by providing new ways of
examining and replaying the field experiences in the VR environments. With the new techniques we have for
building, instrumenting, and collecting field data, the VR environments potentially can allow learner to replay
their own or others' experiences and to experiment with different choices and outcomes. Along with this,
learners can practice in a Virtual Reality environment events or scenarios that are rare or much too dangerous
to do in normal training.
Virtual Reality (VR) and Artificial Intelligence (AI) have emerged as transformative tools in education,
offering immersive and adaptive learning experiences. These technologies can simulate real-world banking
scenarios, such as customer service, financial analysis, and risk management, enabling employees to develop
practical skills in a controlled environment. The integration of VR and AI in learning aligns with Indonesia’s
push toward technological advancement in education and workforce development, particularly in the financial
sector.
According to the 70:20:10 learning model, as outlined by Bob Eichinger and Mike Lombardo in 'The Career
Architect Development Planner' (1996). Learning approach emphasizes:
Experience (70%): Learning and developing by doing, through stretch projects, job rotations, and other hands-
on activities.
Exposure (20%): Learning and developing through others, via coaching, mentoring, and collaborative
interactions.
Education (10%): The most common way to learn and develop, through structured courses and programs such
as webinars, workshops, and e-books. Given that experience constitutes the largest portion of this model, it is
key to maximizing learner development. Therefore, Banks also need to focus on creating learning
opportunities that enable employees to gain real-life experience, ensuring they are well prepared to meet the
challenges in their day-to-day work activities.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025
Page 747
www.rsisinternational.org
Virtual Reality as an Immersive Learning Medium.
Definition of VR Learning. Virtual Reality Learning can be defined as:
“An instructional method that employs computer-generated three-dimensional environments, enabling learners
to experience immersion, interaction, and presence, thereby facilitating experiential and situated learning”
(Freina & Ott, 2015, p. 134).
The three key elements of VR learning are:
Immersion – the sense of being enveloped by a digital environment.
Presence – the psychological perception of “being there.”
Interactivity – the ability to manipulate, respond to, and shape the environment.
For banking education, this means learners can virtually step into a bank branch, interact with digital
customers, analyze transaction flows, and make decisions within a risk-free simulated context.
Virtual Reality, or VR creating an immersive digital environments. VR transports users into computer
generated worlds, completely replacing their existing physical surroundings. For example, you might find
yourself walking through ancient Rome or attending a virtual concert, all while sitting in your living room. VR
provides a completely immersive interaction. Your movements control your virtual body and environment,
isolating yourself from physical distractions.
The concept of VR-based education is rooted in experiential learning theory, which emphasizes “learning by
doing” (Kolb, 1984). Historically, learning in professional domains such as medicine, aviation, and the military
has required simulation-based approaches to reduce risk and enhance preparedness. VR now extends these
benefits to financial education and banking training. A growing body of research highlights VR’s
transformative impact on learning:
•Radianti et al. (2020) conducted a systematic review of VR in higher education, concluding that VR
significantly enhances learner engagement and knowledge retention.
•Freina and Ott (2015) emphasize that VR allows abstract concepts to be experienced concretely, thus
improving conceptual understanding
Virtual Reality (VR) has emerged as a powerful immersive technology capable of supporting experiential and
simulation-based learning. In contrast to traditional digital learning environments, VR enables learners to
interact with realistic, three-dimensional environments, facilitating situated learning and experiential
knowledge acquisition (Radianti et al., 2020). Empirical evidence suggests that immersive VR environments
can enhance learner motivation, engagement, and knowledge transfer when appropriately designed
(Makransky et al., 2020).
Systematic reviews indicate that VR is particularly effective for skills training, decision-making, and complex
problem-solving tasks, which require learners to apply theoretical knowledge in realistic contexts (Parong and
Mayer, 2018). However, prior research also notes that VR alone does not guarantee improved learning
outcomes; its effectiveness depends on instructional design, feedback mechanisms, and learner support
systems (Radianti et al., 2020).
Virtual Reality (VR) provides immersive learning environments that enhance engagement and experiential
learning, while Artificial Intelligence (AI) offers personalized learning experiences through adaptive
algorithms. According to Muhammad Shodiq in ManageHR Magazine, VR learning represents “a new way of
learning” by enabling interactive simulations that mimic real-world scenarios, fostering deeper understanding
and skill development. Shodiq emphasizes that VR can transform traditional training by offering hands-on
practice in a risk-free setting, which is particularly relevant for banking education where practical skills are
critical.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025
Page 748
www.rsisinternational.org
Utilize virtual reality technology able to improve learning scale, scale ability, efficiency and create a better
learning experience for learner. At least, there are five benefits of virtual reality training.
First, Scenario (simulation) based learning. Virtual Reality thru computer-based technology in general has
the potential to greatly impact the ways in which we educate and train all people. VR as a package of computer
hardware, software, and interface technologies that allow a user to experience certain aspects of a simulated 3-
dimensional environment. Upskill employees through real scenario and prepare them for any surprises that
could happen on their job line. Second, Safe training environment for Employee. In this pandemic era,
offline training is not preferred as it could be a media for transmitting virus. Through virtual reality, we
eliminate the need for physical proximity of colleagues, teammates, classmates, or instructors by creating a
distance learning/training environment. This is possible by the technology of networking, simulations, and
interactive software. Because we are able to communicate with others and perform required tasks without the
physical pressure of either fellow performers or actual equipment, we have indeed created a work space
environment which is virtual. VR has the ability to handle the unusual event-and unusual or rare could mean
the need for unusual responses and individual but still in a safe environment. Third, Immersive training.
Digital simulation need to be as realistic as possible to create the illusion of real practice. The degree of
immersion within the synthetic environment as relative to the tasks to be performed and the skill to be
developed by the individual or the group then one is given a wider variety of environments which could be
called virtual. Another interesting thing about these environments is that every object is a state machine.
Therefore, and this again comes back to education and training, the objects that I am holding in my hand, the
rooms I have walked through, the things I've accomplished in that environment will literally determine what I
see and what objects do to me as I walk through this environment. Fourth, VR will save time and money by
allowing us to do virtual training instead of things that usually require field exercises or expert monitoring, or
even things not possible in the field at all. Saves learning time with simulation using practice by doing method.
VR training can reduce training cost by 50- 60 percent. VR training can reduce training time by 40 percent
while improving employee performance by 70 percent. In overall, VR training will increase productivity
through better learning effectiveness and shorter learning life cycle. Fifth, raise learning motivation by
learning using VR and gamification. Virtual Reality is more engaging and will greatly motivate users to
participate in training, or indeed more broadly in education or other activities. The goal of VR environments is
to be very engaging so that the learners will want to learn more and enjoy being part of it. This is a very
important aspect of any kind of system that is trying to support training.
Challenges of VR Learning
Despite its promise, VR adoption in education faces several barriers.
Financial Cost. VR headsets, software licenses, and customized simulation design involve substantial
investment. Small universities or training institutes may struggle with adoption (Radianti et al., 2020).
Technical Limitations. Issues include hardware requirements, high bandwidth needs, and learner discomfort
such as motion sickness. These can undermine the learning experience.
Pedagogical Integration. A key challenge lies in aligning VR content with banking curriculum standards.
Poorly designed VR modules risk becoming entertaining but educationally ineffective (Huang et al., 2020).
User Acceptance and Training. Both instructors and learners may resist adoption due to unfamiliarity with
VR tools. Studies show that successful integration requires capacity building and digital literacy programs.
Ethical and Privacy Concerns. As VR integrates with artificial intelligence (AI) and data analytics, concerns
arise regarding data privacy, surveillance, and equity of access.
Artificial Intelligence in Learning Technologies.
Artificial Intelligence (AI) refers to the simulation of human intelligence processes by machines, particularly
computer systems. These processes include learning - the ability to improve performance based on experience,
reasoning-the ability to make decisions, problem solving, perception-recognizing objects or patterns, language
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025
Page 749
www.rsisinternational.org
understanding, and potentially even creativity. AI systems are typically designed to perform specific tasks or
mimic human behavior in various ways. AI is categorized into two broad types. First, Narrow AI (Weak AI).
This is the most common form of AI and is designed to perform a specific task, such as voice assistants like
Siri or Alexa, recommendation systems, or self-driving cars. Narrow AI does not possess general intelligence
and cannot perform tasks outside its specialized area. Second, General AI (Strong AI). General AI would have
the ability to understand, learn, and apply intelligence across a wide range of tasks, similar to a human being.
This type of AI is still theoretical and has not been achieved yet.
Artificial Intelligence (AI) has emerged as a powerful enabler in reshaping the corporate learning landscape.
Across industries, AI is helping organizations transition from one-size-fits-all training methods to personalized,
adaptive learning ecosystems. In a time when digital transformation is accelerating, there is increasing pressure
for employees to develop relevant skills quickly, efficiently, and in a way that aligns with evolving business
strategies.
Recognizing these dynamics, the need to embrace AI not just as a tool, but as a strategic solution to fill in the
skill gaps, accelerate knowledge transfer, and promote behavior change at scale. AI provides the capabilities to
create:
• Personalized learning paths, where content adjusts based on the learner’s current knowledge or
performance;
• Real-time feedback and support, offering guidance as employees practice new skills;
• Conversational and voice-based learning, simulating natural, human-like dialogues for better
engagement and retention;
• Scalable knowledge delivery, ensuring consistent learning experiences across functions, levels, and
geographies.
The principal application of AI in education is adaptive learning. It employs computation techniques to derive
relevant data from the large volume of data created by learners, including their performance, their interaction,
and their learning style feedback, to recommend and make the necessary changes. This makes learning
flexible; students can advance in class at their rates, relying on their weaknesses. AI also drives natural
language processing (NLP) systems, such as chitchatting and virtual tour tutors, that answer students'
questions and offer timely assistance. Apart from customization, AI also has the potential to perform many
practical assignments like grading and assessment, thus relieving faculty members from several other related
chores.
Artificial Intelligence (AI) has increasingly been adopted in learning technologies to support adaptive learning,
personalized instruction, and data-driven decision-making. Within educational contexts, AI enables learner
modelling, intelligent feedback, and performance prediction through techniques such as intelligent tutoring
systems, learning analytics, and adaptive content delivery (Woolf, 2010; Chen et al., 2020). These capabilities
are particularly relevant in professional and higher education, where competency development and
performance measurement are critical outcomes.
Recent studies highlight that AI-driven learning environments can dynamically adjust learning pathways based
on learner behaviour and performance data, thereby enhancing learning efficiency and effectiveness (Ifenthaler
and Yau, 2020). From an information and learning technology perspective, AI functions not merely as a
computational tool, but as a mechanism for transforming educational data into actionable insights that support
learner engagement and achievement (Holmes et al., 2019).
Integration of AI and VR for Competency and Performance Development
When incorporated, AI and VR make an educational platform most responsive to the users' needs. AI also has
the ability, as a dynamic third party, to monitor learner behaviors in VR environments and make changes to
ensure each learner receives their optimal experience. For example, in a virtual reality chemistry laboratory,
AI can observe how children are conducting their experiments and modify the degree of challenge to
correspond to this fact. The possibilities of how AI can be applied within customized VR learning
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025
Page 750
www.rsisinternational.org
environments are immense, creating opportunities for improved effectiveness, interactivity, and engagement
within learning activities, all of which can be personalized according to learners’ requirements.
Altogether, the synergic use of AI and VR is the new innovative approach to educational technology. In
addition to positive gains in student engagement, incorporating these learning technologies in adaptive learning
driven by artificial intelligence capabilities allows sophisticated, engaging, and effective simulation-based
practice to master knowledge within academic subjects. Alone, the two have the potential to develop education
that is more engaging, inclusive, and fit for purpose in a world that changes with the click of a button
The integration of AI and VR represents a significant advancement in learning technology by combining
immersive experiential learning with intelligent adaptivity and analytics. AI-powered VR environments enable
real-time monitoring of learner behaviour, adaptive feedback, and personalised learning experiences within
immersive simulations (Dede et al., 2017). This convergence addresses key limitations of standalone VR by
embedding intelligence into the learning process.
In professional education domains, such as business, finance, and management, AI-driven VR simulations
have been shown to support competency development by allowing learners to practise decision-making in
realistic, risk-free environments (Faria et al., 2009). AI enhances these environments by analysing learner
interactions and performance, thereby providing tailored feedback and supporting continuous improvement.
From an information and learning technology perspective, AI-powered VR environments also generate rich
learning data that can be utilised for learning analytics, performance assessment, and instructional
improvement (Ifenthaler and Yau, 2020).
Integrating AI technologies into other learning architecture—such as Virtual Reality (VR). By combining
VR and AI, we will create a more interactive, realistic, and humanized experience. AI technologies like
Speech-To-Text (STT), Natural Language Processing (NLP), Text-To-Speech (TTS), and generative models
were embedded to enable fluid two-way conversations between learners and virtual characters. This allows for
more flexible, experiential learning that adjusts in real time to the learner’s input.
This strategic move aligns with organization vision to create learning journeys that are not only scalable and
measurable, but also highly relevant, personalized, and impactful. It also reflects their commitment to
innovation, digitalization, and employee development in a future-ready organization.
By embedding responsible AI principles into the core of organization development process, they ensured that
innovation is accompanied by accountability. This foundation has allowed they build a learning experience that
employees can trust and that the organization can scale with confidence.
AI principles and governance practices to ensure the safety, fairness, and integrity of the solutions.
Key ethical foundations in AI development include:
1. Transparency & Explainability. All AI interactions are designed to be understandable by users. Learners
are aware when they are interacting with AI, and the learning system provides clear feedback and
guidance to support the development process.
2. Data Privacy & Security. The process followed strict internal data management protocols. All learner
data used for testing or improvement purposes was anonymized and securely stored in compliance with
internal risk policies and regulatory requirements.
3. Bias Mitigation. Training content and conversational responses were reviewed to avoid reinforcing
stereotypes or favoring specific learner profiles. AI-generated responses were tested across diverse
inputs to ensure fairness.
4. Governance Compliance. The solution was developed under the oversight of AI Governance
Framework, ensuring that all components went through necessary risk, IT, and compliance reviews
before implementation.
5. Human Oversight & Iterative Testing. Organization maintained strong human involvement throughout
the development process—from model training and scenario design to user testing and content
validation. This ensured the AI served as an enabler, not a replacement, for meaningful human learning.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025
Page 751
www.rsisinternational.org
Competency-based learning frameworks
A competency framework serves as a foundational tool for talent development in banking. It defines the
essential skills, behaviors, and knowledge required for employees to perform effectively.
Competency in banking includes technical knowledge, problem-solving skills, and interpersonal abilities.
Performance outcomes can be measured through task accuracy, efficiency, and customer satisfaction. The
importance of learning strategies that prioritize active engagement and assessment, which VR and AI can
facilitate through real-time feedback and scenario-based learning.
Training Evaluation Model
The Kirkpatrick Model for Training Evaluation. Training evaluation is a critical component of human resource
development, aimed at assessing the effectiveness, efficiency, and impact of training programs. Among various
models, the Kirkpatrick Four-Level Model, introduced by Donald Kirkpatrick in 1959, remains the most
widely used framework for evaluating training outcomes across industries.
Overview of the Kirkpatrick Model
Kirkpatrick’s model evaluates training effectiveness through four progressive levels:
Level 1: Reaction. Measures participants’ immediate responses to the training, including satisfaction and
engagement. Positive reactions are often linked to increased motivation and learning.
Level 2: Learning. Assesses the extent to which participants acquire intended knowledge, skills, and attitudes
during the training. This level often involves pre- and post-tests or assessments.
Level 3: Behavior. Evaluates the transfer of learning to the workplace. It examines whether participants apply
what they learned in their job roles, typically through observation or feedback from supervisors.
Level 4: Results. Measures the final impact of training on organizational outcomes, such as productivity,
quality, customer satisfaction, or financial performance. This level is the most complex and resource-intensive
to evaluate.
A comprehensive learning strategy ensure employees stay competent, compliant, and adaptable in banking
industry that is constantly evolving. Bank is highly regulated industry, and non-compliance can lead to severe
penalties. A well-structured learning architecture helps employees stay updated with regulatory changes, risk
management practices, and financial technologies, reducing operational and reputational risks.
AI Powered VR Learning Development
AI VR Learning Solution Architecture & Workflow
Overview Process Flow
The user's journey through a training session follows a seamless, five-stage process powered by a sophisticated
AI backend. This ensures that every interaction is both realistic and purposeful.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025
Page 752
www.rsisinternational.org
Detailed Step-by-Step Workflow
Here is a granular breakdown of each stage in the technical flow.
Step 1: User Speech Input & Transcription
Technical Flow: Microphone > Whisper STT (Speech-to-Text)
Explanation: The employee speaks naturally into the VR headset's microphone. The audio is instantly captured
and processed by OpenAI's Whisper, a state-of-the-art transcription engine, which converts the speech into a
highly accurate text transcript. This forms the foundation for the AI's understanding.
Step 2: Input Sanitization & Contextualization
Technical Flow: AI Agent Glossary + LLM for Sanitization
Explanation: To bridge the gap between natural human speech and structured system input, the raw transcript
is "sanitized." For instance, due to dialect, pronunciation, or ambient noise, a user might say "CIMB," but the
initial transcript could read "Jambi" or "Simbi." The sanitization process, guided by our Elwyn Agent that
consists of module-specific glossary, automatically corrects such discrepancies. This critical step ensures the
conversation's core context is always preserved and accurately understood by the system.
Step 3: Scenario-Driven Response Generation
Technical Flow: AI Agent Scenario + LLM
Explanation: Once the user's intent is clear, the AI Agent provides the necessary context to the LLM. The
LLM then generates a response that is not only contextually appropriate but is also strictly aligned with the
pre-defined learning objectives of the specific training scenario. This ensures that every AI interaction
purposefully guides the employee toward the desired learning outcome.
Step 4: Multi-Modal AI Response Delivery
Technical Flow: AI Response (Visual Text + ElevenLabs TTS + Animation Trigger)
Explanation: The AI's response is delivered through three synchronized channels to maximize immersion:
Visual Text: The reply appears on-screen for clarity.
Natural Speech: Using the industry-leading ElevenLabs TTS engine, the text is converted into remarkably
human-like audio.
Avatar Animation: Our VR application triggers one of several predefined character animations (e.g., nodding,
looking confused, smiling). The choice of animation is based on a suggestion from the GenAI, which analyzes
the tone and content of its own response to recommend the most fitting non-verbal cue.
Step 5: Objective, Rubric-Based Performance Analysis
Technical Flow: AI Agent Score Handler + LLM
Explanation: Upon session completion, the system provides a comprehensive evaluation based on a scoring
rubric co-developed with the relevant CIMB business unit. This ensures that even though the conversation is
dynamic and unscripted, the final assessment remains objective and standardized. This approach makes
dynamic learning effectively measurable, tying free-form practice directly to concrete performance metrics
agreed upon at the project's outset.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025
Page 753
www.rsisinternational.org
Research Methods
This study employed a convergent parallel mixed-methods design, combining quantitative and qualitative
approaches to evaluate the effectiveness of AI-powered VR learning in wealth management education. The
choice of this design aligns with Creswell and Plano Clark (2018), who argue that integrating both types of
data provides a more comprehensive understanding of educational interventions. The approach also supports
triangulation, enhancing the reliability and validity of findings.
The adapted model consists of four levels, with two of them (reactions and learning) evaluated by quantitative
data (a survey) with learners. The behaviour level was evaluated by qualitative and quantitative data (open-
and closed-ended questions) with learners because a quantitative study may not be sufficiently flexible to
facilitate an understanding of complex human behaviour or to address sensitive matters [Johnson, R.B.;
Onwuegbuzie, 2004]. The results level was evaluated by business impacts such MLD booking volume, cost
avoidance as well as the operational risks result.
Level 1: Reaction. Measures participants’ immediate responses to the training, including satisfaction and
engagement. Positive reactions are often linked to increased motivation and learning.
There are 6 (six) questions related to Level 1 evaluation:
1. The content in the VR learning module aligns with the learning objectives
2. The VR learning module enhances my effectiveness in carrying out my responsibilities
3. The instructional content delivered through the VR learning module is clearly structured and easily
comprehensible.
4. The VR headset is user-friendly and facilitates an effective learning experience
5. VR technology enables highly responsive interactions that actively engage users throughout the
learning experience
6. The visual and audio components of the VR learning experience are of high clarity and sufficiently
meet instructional standards
The Level 1 score scale out of 5 points
Level 2: Learning. Assesses the extent to which participants acquire intended knowledge, skills, and attitudes
during VR learning modules especially on MLD product complaint handling & VR AI – MLD Product. The
Level 2 score scale out of 100 points.
Level 3: Behavior. Evaluates the transfer of learning to the workplace. It examines whether participants apply
what they learned in their job roles:
The knowledge and skills acquired through the VR learning module have been effectively applied in my
day-to-day professional responsibilities.
The Level 3 score scale out of 5 points
Level 4: Results. Measures the final impact of VR learning module MLD product on organizational outcomes,
such as MLD volume booking, training cost avoidance & reduction of operational risks especially miss selling
risks.
RESEARCH RESULTS
The integration of AI-powered Virtual Reality (VR) learning environments is significantly transforming
competency development in wealth management education.
Survey Periode : 15-31 August 2025
Participants: 2.608 staff
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025
Page 754
www.rsisinternational.org
A total of 2,608 staff from PT. Bank CIMB Niaga, Tbk participated, including relationship managers, product
managers, and branch-level staff. Participants were selected using purposive sampling to ensure inclusion of
employees directly involved in wealth management advisory.
Quantitative Data: Collected through Likert-scale surveys (Levels 1–4 of Kirkpatrick’s model) consist of
reaction evaluation, knowledge & performance assessments, behavioral change as well as business impacts.
Metrics included knowledge acquisition, decision-making speed, accuracy, and satisfaction levels.
Qualitative Data: Gathered from open-ended survey questions, semi-structured interviews, and focus group
discussions. This enabled deeper exploration of participants’ experiences, attitudes, and perceptions of AI-VR
training.
Quantitative Analysis: Descriptive statistics, paired-sample t-tests, and ANOVA were used to evaluate
improvements in competencies and compare pre- and post-training performance. Reliability of survey
instruments was tested using Cronbach’s alpha (>0.7 considered acceptable).
Qualitative Analysis: Thematic analysis (Braun & Clarke, 2006) was applied to identify recurring themes, such
as enhanced confidence in client communication, improved comprehension of structured products, and
motivation for continuous learning.
Level 1: Reaction
1
The content in the VR learning module aligns with the learning objectives
4.94
2
The VR learning module enhances my effectiveness in carrying out my responsibilities
4.94
3
The instructional content delivered through the VR learning module is clearly structured and easily
comprehensible.
4.94
4
The VR headset is user-friendly and facilitates an effective learning experience
4.94
5
VR technology enables highly responsive interactions that actively engage users throughout the
learning experience
4.93
6
The visual and audio components of the VR learning experience are of high clarity and sufficiently
meet instructional standards
4.94
Level 1 (Reaction): Participants reported extremely positive reactions to the VR training, with an average
score of 4.94/5, indicating strong satisfaction and engagement.
Level 2: Learning
AI tracks metrics like:
Accuracy and decision-making speed.
Engagement levels and completion rates.
Behavioral patterns and learning progression.
This enables:
Immediate feedback and course correction.
Predictive analytics to forecast future performance.
Benchmarking against peers for motivation and self-assessment
1
MLD Product Complaint Handling
95.5
2
VR AI: MLD Product
84.0
Average L2 Score
89.8
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025
Page 755
www.rsisinternational.org
Level 2 (Learning):Competency acquisition improved significantly, with average scores reaching 89.8/100,
particularly in product complaint handling (95.5/100). Statistical tests (paired-sample t-test, *p* < 0.01)
confirmed the significance of learning improvements compared to pre-training levels.
Level 3: Behavior
The knowledge and skills acquired through the VR learning module have been effectively applied in
my day-to-day professional responsibilities
4.14
Level 3 (Behavior):Participants demonstrated effective transfer of learning, scoring 4.14/5on applying new
skills in daily work.
Open-ended questions:
Around 40–50% of the feedback highlighted participants' ability to apply, implement, or feel more prepared to
integrate the learning outcomes into their day-to-day professional responsibilities.
Level 4: Results.
Incorporating AI-powered VR into wealth management training can simulate complex financial scenarios,
enhance regulatory and compliance understanding, improve advisory skills through realistic client interactions
and support continuous professional development with adaptive learning modules.
MLD Training Cost Efficiency Ratio: 92.18%
MLD operational risks related to miss selling HoH from 25% to 0%
Level 4 (Results):Business outcomes showed measurable improvements:
Incremental MLD booking volume increased by IDR 25.5 billion HoH.
Training cost efficiency reached 92.18% , reducing dependency on external trainers as well inhouse training
program
Operational risks of mis-selling dropped to 0%
Thematic analysis of open-ended responses revealed three dominant themes:
1. Increased confidence in client communication and advisory tasks.
2. Improved comprehension of complex structured products like MLDs.
3. Higher motivation to engage in digital and self-directed learning.
Illustrative participant feedback included statements such as: “The VR simulation felt like I was in a real client
meeting, which helped me practice handling objections without fear of mistakes.”and “I now feel more
confident explaining risks of structured products.”
The findings are consistent with Hamilton et al. (2021), who reported improved retention and engagement with
VR learning in technical fields, and Ifenthaler et al. (2024), who emphasized AI’s role in adaptive feedback.
However, this study contributes uniquely by linking VR-AI learning directly to business outcomes (booking
volume, risk reduction), which has been underexplored in prior research.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025
Page 756
www.rsisinternational.org
DISCUSSION & CONCLUSION
This study adapted the Kirkpatrick evaluation model, which includes four levels of training outcomes, to
evaluate AI powered VR learning in enhancing competency & performance in wealth management education.
Based on L1 evaluation, learners perceive the effectiveness and engagement of AI-VR-based training
compared to traditional methods with average score 4.94 (out of 5)
Based on the L2 assesment, AI-powered VR learning environments affect competency development in wealth
management education with average score 89.8 out of 100.
Based on the L3 : 50% of the feedback highlighted participants' ability to apply, implement, or feel more
prepared to integrate the learning outcomes into their day-to-day professional responsibilities with average
score 4.14 (out of 5)
Based on L3 & L4 Learners exposed to AI-powered VR simulations demonstrated significantly higher
competency in wealth management advisory tasks than compared before implementing AI -powered VR
learning through conventional methods shown by the incremental of MLD booking volume HoH by IDR 25.5
billion, training cost avoidance by 92.18% as well as reducing operational risks HoH from 25% to 0%
This study provides empirical evidence that AI-powered VR training significantly enhances competency and
performance in wealth management education in Indonesia.
Q1: Learners exposed to VR-AI training demonstrated higher competency in advisory tasks, supported by
strong improvements in knowledge scores and behavioral application.
Q2:The immersive environment improved learners’ understanding of complex financial products (e.g., MLDs)
and advisory performance, confirmed by both quantitative and qualitative findings.
RQ3: Participants reported high engagement and satisfaction compared to traditional methods, validating the
superiority of AI-VR over classroom-based approaches.
The study extends existing literature by demonstrating that beyond improving knowledge and skills, AI-VR
training can generate measurable business outcomes, such as increased product sales and reduced compliance
risks.
LIMITATIONS
The study was limited to one banking institution, restricting generalizability. It focused on short-term outcomes
(August 2025), with limited data on long-term retention. Adoption requires investment in VR infrastructure,
which may challenge scalability in resource-constrained institutions.
RECOMMENDATIONS
For Banking Practice
1. Integration into Continuous Professional Development (CPD): AI-VR learning should be embedded
into continuous professional development for wealth managers, expanding coverage to include
insurance, investment funds, and advanced risk profiling.
2. Real-time Analytics: Banks should leverage AI dashboards to monitor staff competencies and adjust
training dynamically.
3. Scalability: Invest in VR hardware-as-a-service or shared VR labs to reduce infrastructure costs.
For Regulators (OJK)
1. Develop certification frameworks incorporating VR-based simulations for structured product advisory.
2. Incentivize adoption of AI-VR solutions by recognizing them as tools for reducing mis-selling risks and
improving compliance.
INTERNATIONAL JOURNAL OF RESEARCH AND INNOVATION IN SOCIAL SCIENCE (IJRISS)
ISSN No. 2454-6186 | DOI: 10.47772/IJRISS | Volume IX Issue XII December 2025
Page 757
www.rsisinternational.org
For Future Research
1. Conduct longitudinal studies to assess knowledge retention and behavior sustainability over 6–12
months.
2. Compare AI-VR vs. AI-only vs. VR-only training interventions to isolate their unique effects.
3. Undertake cost-benefit analyses to evaluate scalability across multiple financial institutions.
REFERENCES
1. Bauer, E., Greiff, S., Graesser, A.C., et al. (2025). Looking Beyond the Hype: Understanding the
Effects of AI on Learning. Educational Psychology Review, 37, Article 45.
https://doi.org/10.1007/s10648-025-10020-8[2](https://link.springer.com/article/10.1007/s10648-025-
10020-8).
2. Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in
Psychology, 3(2), 77–101.
3. Creswell, J. W., & Plano Clark, V. L. (2018). Designing and Conducting Mixed Methods Research
4. Gunardi, H., Primiana, I., Effendi, N., Herwany, A. (2024). The Performance of Private Wealth
Management in Indonesia. Journal of the Academy of Banking Studies, 24(1), 634. Link.abacademies
5. Hamilton, D., McKechnie, J., Edgerton, E., Wilson, C. (2021). Immersive Virtual Reality as a
Pedagogical Tool in Education: A Systematic Literature Review. Journal of Computers in Education, 8,
1–32. https://doi.org/10.1007/s40692-020-00169-2[3](https://link.springer.com/article/10.1007/s40692-
020-00169-2).
6. Ifenthaler, D., Majumdar, R., Gorissen, P., et al. (2024). Artificial Intelligence in Education:
Implications for Policymakers, Researchers, and Practitioners. Technology, Knowledge and Learning,
29, 1693–1710. https://doi.org/10.1007/s10758-024-09747-
0[1](https://link.springer.com/article/10.1007/s10758-024-09747-0).
7. Johnson, R.B.; Onwuegbuzie, A.J. Mixed methods research: A research paradigm whose time has
come. Educ. Res. 2004, 33, 14–26
8. Senduk, W.F., Djatmika, E.T., Wahyono, H., et al. (2024). Fostering Financially Savvy Generations:
The Intersection of Financial Education, Digital Financial Misconception and Parental Wellbeing.
Frontiers in Education,
9. https://doi.org/10.3389/feduc.2024.1460374[3](https://www.frontiersin.org/journals/education/article
s/10.3389/feduc.2024.1460374/full).
9. Shodiq, M. (2024). Virtual Reality (VR) learning —a new way of learning. ManageHR Magazine.
10. Shodiq,M.(2025). AI Powered VR Learning. ManageHR Magazine. https://corporate-online-training-
apac.managehrmagazineapac.com/cxoinsight/artificial-intelligence-ai-powered-virtual-reality-vr-
learning-nwid-4272.html
11. Smallwood, M., Prosowicz, G., Dirga, A., et al. (2024). Wealth Management in Indonesia – Enormous
Opportunities & Great Challenges. Hubbis. PDF Link.hubbis
