Category: Knowledge

Introduction: The Institutional Gateway to Digital Assets and the Trust Imperative

Blockchain technology has matured beyond its origins. It now powers a diverse universe of digital assets, poised to unlock unprecedented institutional opportunities and drive business innovation. From alternative investment like ETFs and ETPs to stablecoin payments and efficiency-enhancing real-world asset tokenization e.g. bonds, equity, private debt, the institutional allure is undeniable.

This surge in institutional interest is fueled by several tailwinds: greater regulatory clarity, a maturing ecosystem of service providers such as custody, trading infrastructure, compliance solutions, data analytics, risk management tools, and rising client demand for yield. However, the linchpin for widespread and sustainable adoption lies in establishing robust trust – trust among market participants, regulators, and investors. This trust is built upon a foundation of comprehensive risk management and cutting-edge technology.

Recent event underscores this trust imperative. The Bybit heist, with approximately $1.46 billion in stolen crypto assets, demonstrates that even crypto-native companies are vulnerable to sophisticated attacks like supply chain exploits, UI manipulation, and social engineering. These complexities pose significant challenges for institutions, particularly those new to the digital asset space.

At Beacon VC, we believe that robust risk frameworks and advanced technologies are essential prerequisites for secure institutional engagement with digital assets. This conviction shapes our investment thesis: identifying opportunities in the infrastructure, tools, and services that empower this trust-building process.

In the article, we will walk through key vulnerabilities for institutions entering digital asset space, guideline on how to build comprehensive risk frameworks, technology landscape for digital asset compliance solutions, and how building these two foundations can create trust internally and externally.

Navigating the Labyrinth: Key Vulnerabilities for Institutions

Institutions venturing into digital assets face a spectrum of unique risks, stemming from both internal and external sources. These risks are amplified by the nascent nature of the digital asset class, characterized by challenges such as insufficient investor education, regulatory frameworks struggling to keep pace with the evolving digital asset landscape, and developing infrastructure, which collectively contribute to heightened fraud, market manipulation, money laundering, and overall market instability.  This is further compounded by regulatory uncertainty and the borderless nature of digital assets, which facilitate cross-border threats and often outpaces businesses’ ability to protect investors. These internal and external vulnerabilities give rise to the following specific risks for institutions.

• Internal Vulnerabilities: These vulnerabilities arise from within the institution’s own technology, operations, and human factors.
  • These relate to the systems and infrastructure used to manage digital assets.
    • Private Key Compromise: can result in irreversible asset losses. Examples: supply chain attacks (third-party vendor), phishing, malware.
    • Cybersecurity Breaches: Weaknesses in cybersecurity defenses can be exploited. Examples: network intrusions, ransomware attacks, DDoS attacks.
    • Smart Contract Vulnerabilities: Flaws in smart contract code can be manipulated. Examples: code exploits, flash loan attacks.

• • Human-Centric Risks: These vulnerabilities stem from human error, lack of awareness, or inadequate control.
    • Lack of Adequate Training/Awareness: A deficient understanding of digital asset risks exposes institutions to increased operational and security risks due to employee errors.
    • Weak Internal Controls/Governance: Insufficient controls and governance create opportunities for errors, fraud, or unauthorized activities.
    • Operational Inefficiencies/Errors: Inefficient processes and human error in managing digital asset operations e.g., manual processes for transactions, reconciliation, or reporting can lead to losses and increased risks.  
    • Absence of Clear Policies/Procedures: Lack of well-defined internal policies for digital asset activities e.g., absence of policies for acceptable use, incident response, or employee trading can increase vulnerabilities and inconsistencies.

• External Vulnerabilities: These vulnerabilities arise from factors outside the institution’s direct control.
  • Regulatory Risk: Failure to comply with regulations can result in penalties and legal issues such as deficiencies in KYC/AML controls. Furthermore, the inherent uncertainty in the evolving regulatory framework for digital assets creates a risk that any related decision may become non-compliant.
  • Counterparty Risk: Risks associated with entities with which an institution interacts in the digital asset space, including exchanges, custodians, DeFi protocols, and other financial institutions. This risk encompasses the potential for these entities to default on their obligations, experience financial distress, or suffer operational failures.
  • Protocol and Smart Contract Risks: technology or DeFi protocols that an institution uses. This includes risks from vulnerabilities in the underlying blockchain consensus mechanisms, potential for forks or network splits, and exploits of flaws in the code of smart contracts, which can lead to loss of funds or disruption of operations.
  • Market-Related Risks:
    • Market Manipulation and Fraud: Digital asset markets’ novelty and, in some cases, lack of regulation make them susceptible to manipulation e.g., pump-and-dump schemes and fraud.
    • Geopolitical and Systemic Risks: External events e.g., government actions, network outages, significant market events can impact on the digital asset market. For instance, geopolitical tensions can lead to divergent and conflicting regulations across jurisdictions, creating a complex operating environment for institutions, while state-sponsored cyberattacks could target critical digital asset infrastructure, leading to systemic risks. These external factors introduce a layer of uncertainty that institutions must be prepared to navigate.

Considering these internal and external vulnerabilities, institutions face not only the risk of direct financial losses but also significant reputational risks. Both internal failures and external associations can severely harm an institution’s standing. Therefore, the next crucial step is to establish robust risk frameworks and leverage technology to proactively address and mitigate these potential financial and reputational risks.

The Cornerstones of Trust: Comprehensive Risk Frameworks

Institutions venturing into digital assets must prioritize robust risk frameworks that encompass governance, security, compliance, market risks, and operational aspects. These frameworks are essential not only for regulatory compliance, but also for sound internal operations.

Frameworks such as the Digital Asset Security Control Practices (DASCP) framework offer a valuable baseline for the industry. The DASCP is highlighted because its comprehensive and adaptable design is particularly well-suited to the digital asset space, enabling institutions to navigate current challenges and foster an inclusive resilient financial ecosystem that can evolve alongside technological advancements. DASCP employs a layered approach, establishing foundational principles, identifying associated risks, and designing flexible controls to address those risks.

The principles cover critical areas as the following from top priority as a necessary requirement to subsequent priority to achieve full potential.

• Legal Certainty: Ensure operations comply with current and evolving legal frameworks
• Regulatory Compliance: Align processes, controls, and procedures with specific rules and regulations issued by relevant regulatory bodies
• Resilience and Security: Build robust infrastructure and processes to withstand disruptions and protect data/assets
• Safeguarding Customer Assets: Establish strong governance and controls to securely manage client holdings
• Connectivity and Interoperability: Enable seamless transactions and settlements across diverse networks for efficient operations
• Operational Scalability: Design efficient, cost-effective systems through standardization and automation to handle increasing volume

Ultimately, institutions must tailor their risk management strategies to their specific circumstances, considering factors such as governance structure, business partner and service provider evaluation, data acquisition and analytics, transaction monitoring, and threat monitoring.

The Technological Imperative: Building Secure Foundations

In the preceding section, the necessity of comprehensive risk frameworks is emphasized to guide institutions in navigating the complexities of digital assets. These frameworks provide blueprints, and technology provides the necessary tools and infrastructure to execute that blueprint, enabling practical implementation and automation of the frameworks. According to Research and Markets, the Regulation Technology (RegTech) market is experiencing significant growth, projected to rise from US$ 7.55 billion in 2023 to US$ 42.73 billion by 2031, with a CAGR of 24.2%. This growth rate is likely even higher for the digital asset-specific RegTech segment, driven by increasing regulatory scrutiny, institutional adoption, the complexity of digital assets, the rise in illicit activities, and the demand for transparency and trust, necessitating specialized solutions for compliance, risk management, and security. As digital assets move toward mainstream adoption, RegTech plays a pivotal role in this transformation, leveraging technologies such as machine learning, AI, natural language processing, blockchain to bring digital transformation to compliance, focusing on areas like blockchain analytics, smart contract verification, risk intelligence, streamlined reporting, and combining advanced algorithms with human oversight.

In response to these evolving technological demands, digital asset compliance solutions can be categorized based on the fundamental requirements and challenges institutions faced when engaging with digital assets, ensuring security and adhering to regulatory obligations. Building upon the discussion of internal and external vulnerabilities and the importance of risk frameworks, this categorization demonstrates how technology provides practical tools to manage those risks. Specifically, custody solutions address private key compromise, cybersecurity solutions defend against cyberattacks, blockchain analytics and forensic tools help combat money laundering and fraud, and compliance automation enables compliance with regulatory demands. It follows logical progression from the foundational need for secure asset storage to the ongoing requirements for monitoring, risk management, and automated compliance processes.

These four categories represent the core pillars of a robust digital asset compliance framework for institutions, each addressing a fundamental requirement for security, regulatory adherence, and operational efficiency, collectively contributing to building the necessary trust for wider institutional adoption of digital assets.

Securing Trust: Implementing Risk Management and Technology

Building trust, both internally and externally, is crucial for the successful adoption and integration of digital assets by institutions. Internally, robust risk frameworks and secure technologies are essential to foster confidence among leadership, employees, and shareholders, facilitating internal buy-in by demonstrating effective risk management and secure asset handling. A well-trained and informed workforce is also vital for security and compliance.

Externally, institutions must demonstrate a proactive and responsible approach to build trust with stakeholders. This involves assuring clients of the safety of their digital assets through robust custody solutions and transparent reporting, providing regulators with assurance of compliance and market integrity via proactive engagement and adherence to evolving standards, and establishing credibility in the broader market through publicly share their security protocols and certifications or undergo independent audits and attestations to provide third-party verification of their security and compliance measures. A strong security and compliance posture mitigates reputational risk and attracts talent and partnerships by signaling a commitment to best practices.

Conclusion: Driving the Future of Institutional Digital Asset Adoption

The path to widespread institutional adoption of digital assets is paved with trust. This trust is not a given; it must be earned through the diligent implementation of robust risk management frameworks and the strategic deployment of advanced technologies. By proactively addressing vulnerabilities and prioritizing security, compliance, and operational efficiency, institutions can unlock the transformative potential of digital assets. Looking ahead, as trends like MiCA, institutional DeFi, and real-world asset tokenization reshape the landscape, the need for trust will only intensify. Beacon VC recognizes the pivotal role of companies developing the trust infrastructure necessary for this evolution, and we are committed to supporting their growth and innovation.

Building upon this concept of trust, the question, “Beyond Trust: The Next Frontier for Institutional Digital Asset Strategies,” warrants careful consideration to further explore the potential of institutional digital assets.

Author: Wanwares Boonkong (Pin)

Editors: Supamas Bunmee (Jae), Woraphot Kingkawkantong (Ping)

References:

https://medium.com/riva-markets/the-top-30-global-banks-digital-assets-use-cases-756e30f4b451

https://www.elliptic.co/blog/bybit-hack-largest-in-history

https://www.csis.org/analysis/bybit-heist-and-future-us-crypto-regulation#:~:text=On%20February%2021%2C%202025%2C%20a,%241.5%20billion%20in%20Ethereum%20tokens

https://www2.deloitte.com/us/en/pages/advisory/articles/crypto-digital-asset-risk-management.html

https://web-assets.bcg.com/8f/e0/0364f7cb482381051547be407437/dtcc-digital-assets-dascp-whitepaper-design-final.pdf

https://eflowglobal.com/how-regtech-is-shaping-the-future-of-crypto-compliance/

https://www.rblt.com/fintech-insights/regtech-issues-in-cryptomarkets-and-digital-assets

https://www.globenewswire.com/news-release/2025/01/23/3014377/28124/en/Regtech-Market-Size-Regional-Share-Forecast-and-Growth-Opportunity-Analysis-with-Profiles-of-IBM-Deloitte-Thomson-Reuters-PWC-MetricStream-Jumio-ACTICO-Acuity-Ascent-Technologies-a.html

https://cointelegraph.com/explained/ddos-attacks-in-blockchain-networks-explained

https://www.aon.com/en/insights/cyber-labs/flash-loan-attacks-a-case-study

https://coinmarketcap.com/academy/glossary/supply-chain-attack

https://cryptoslate.com/2021-was-the-year-of-institutional-crypto-adoption/

Artificial intelligence is rapidly evolving, moving beyond simple automation to create truly intelligent systems capable of independent action.This evolution has led to the emergence of the AI agent, a sophisticated form of AI designed to perceive its environment, make decisions, and act autonomously to achieve specific goals with minimal human intervention. This article explores the journey of AI from rule-based automation to the dawn of agentic AI, examining its evolution, distinguishing characteristics, operational mechanisms, diverse applications, and the challenges and opportunities that lie ahead in its widespread adoption.

What is AI Agent and how has agentic AI evolved?

Agentic AI or AI agent refers to AI applications designed to function and make decisions to reach specific goals, under changing environments, autonomously with minimal human intervention. To gain a deeper understanding of Agentic AI, it is essential to explore its evolution, its differences from other AI applications, how it works, and its diverse use cases.

Although there are some controversies around whether the terms “Agentic AI” and “AI agent” can be used interchangeably[1], for the purpose of this article, “Agentic AI” is considered analogous to the brain whereas AI agents are seen as the hands-taking action as the brain commands.

Agentic AI evolution and differences from other AI applications

The evolution of Agentic AI can be traced back to early robotic process automation (RPA), which relied on predefined rules and workflows. These systems excelled in environments with clear parameters and predictable conditions, performing structured and repetitive tasks efficiently. However, advancements in large language models (LLMs) have significantly expanded the capabilities of AI systems, resulting in the introduction of conversational AI. Conversational AI is capable of understanding and responding to human languages. Early conversational AI systems, such as chatbots, provided scripted responses within predefined domains.

The integration of Conversational AI with RPA gave rise to AI copilots that could think and act beyond predefined rules. These AI copilots understand natural language, adapt to dynamic environments, and perform more complex tasks. They offer contextual assistance and support human decision-making. While they enhance productivity by interpreting complex inputs and providing intelligent suggestions, they ultimately rely on human direction to guide their actions. As AI models become more sophisticated, automation evolves to handle more complexity and achieve greater autonomy. This progression will culminate in the emergence of Agentic AI—fully autonomous systems capable of tackling sophisticated tasks with minimal human oversight. Currently, various use cases illustrate an intermediate stage between AI copilots and fully autonomous AI agents, where AI systems exhibit greater independence while still requiring some human intervention.

Note: While both RPA and AI agents are well-suited for high-volume, repetitive tasks, RPA is generally limited to tasks with predictable, single outcomes triggered by specific conditions. AI agents, on the other hand, can handle tasks with multiple potential outcomes, adapting and adjusting their approach based on the current situation.

How it works

To better illustrate how Agentic AI could be applied to real-life workflows, take loan assessment as an example. In a normal day, when a relationship manager (RM) receives a loan application from a borrower, they gather data such as financial statements, commercial agreements, and credit bureau data from multiple sources. The RM then collaborates with credit analysts to analyze this data, create a credit approval memo, and seek approval from an authorized person. This process can take up to three weeks or longer.

With agentic AI, the RM’s role would shift from data collection, collaboration with credit analysts, and seeking approvals to simply providing prompts to the AI. The agentic AI would function as a virtual employee, handling all the tasks and reporting the final result back to the RM. This result could be an email to the borrower rejecting the loan or a confirmation email approving the loan with next-step instructions.

To reach a decision, the AI agent would break down the process into subtasks, assigning them to specialized AI agents. For example, an AI data collection agent would gather information from various sources independently, an AI analyst agent would assess creditworthiness and repayment capabilities, and an AI memo agent would compile the analysis and create a credit memo for review by the RM and analysts.

The agentic AI would transparently demonstrate its decision-making process for validation by the RM. If the RM trusts the AI’s reasoning, the RM can authorize the AI to proceed. Thus, the RM’s role evolves from performing all tasks to becoming a validator. This human validation is crucial during the AI’s development phase. However, as the AI’s reliability increases, the RM may eventually only need to oversee the process, allowing the AI to proceed without intervention.

Benefits of Agentic AI

AI agents are rapidly transforming the way businesses operate and interact with their customers. By automating processes, enhancing decision-making, and personalizing experiences, these intelligent systems are driving significant improvements across various sectors. The following sections explore key benefits of AI agent implementation, showcasing real-world examples of how these technologies are boosting efficiency and productivity, accelerating and enhancing decision-making, and ultimately, improving customer experience.

• • Boosting Efficiency and Productivity

AI agents can significantly boost efficiency and productivity by streamlining business processes and automating unstructured tasks. For example, in customer service, AI agents can resolve customer issues by understanding their pain points, gathering relevant information from databases (including historical data), and taking appropriate action. By swiftly completing tasks that typically consume significant human resources, AI agents free up staff to focus on more strategic and meaningful work, ultimately leading to improved overall productivity.  Agentic AI goes beyond the capabilities of AI copilots. While Agentic AI can make autonomous decisions and take appropriate actions, AI copilots function alongside humans, assisting but not independently deciding or acting.

Real-world use cases

• • • • Amazon’s agentic AI (“Amazon’s Warehouse Robots”), manages inventory, predicts demand, and optimizes delivery routes in real-time. These robots navigate complex warehouse environments, adapt to changing conditions, and autonomously transport goods. By leveraging agentic AI, Amazon not only replaces human workers with robots capable of precise task execution, freeing up staff for more valuable work, but also reduces costs associated with scaling its business and hiring additional personnel.[2]
      • PepsiCo leverages agentic AI to streamline recruitment by matching candidates to job roles. The AI scans profiles across multiple sources to generate tailored candidate lists for each position. This enhances productivity, allowing HR teams to focus on higher-value tasks.[3]
  • Accelerating and Enhancing Decision-Making

AI agents excel at enhancing decision-making by quickly and accurately analyzing large volumes of data. They can provide rapid analysis of complex scenarios, accelerating the decision-making process. This is particularly valuable in time-sensitive situations where quick action is crucial. They can also simulate different scenarios and predict potential outcomes, allowing decision-makers to evaluate options quickly.

Real-world use case

• • • • Darktrace’s Enterprise Immune System leverages AI to learn an organization’s typical network behavior. Upon detecting anomalous activity, such as unusual logins or data transfers, the system autonomously blocks threats or isolates compromised devices, effectively halting attacks before they can spread. Relying on manual human review for such tasks is inherently slower and more prone to error, preventing the real-time action needed to minimize potential losses.

• • Improving Customer Experience

AI agents can process vast amounts of data to deliver personalized recommendations based on each customer’s historical data, thereby improving satisfaction and loyalty. Previously, businesses struggled to customize recommendations for individual clients due to high costs, limited data, and time constraints. However, with AI agents, businesses can gain deep insights into customer preferences and provide tailored products or services in real-time. This capability enhances the overall customer experience and strengthens customer relationships. As AI agents achieve greater reliability and earn user trust, they will become a valuable virtual workforce for supporting customers.

Real-world use case

• • • • A digital health company, Livongo, has integrated Agentic AI into its diabetes management system. The AI autonomously analyzes continuous glucose monitoring (CGM) data, dietary habits, and physical activity metrics to generate personalized recommendations and actionable alerts. Instead of merely presenting data, it advises users on actions like consuming carbohydrates when glucose levels trend toward hypoglycemia. This not only saves cost and time for patients who would otherwise need frequent hospital visits and endure long queues but also improves their quality of life by providing real-time recommendations and enabling immediate action, resulting in a better customer experience. [4]
      • A leading Dutch insurer has integrated AI agents into its claims management system, reducing processing time by 46% and increasing customer satisfaction by 9%. Upon receiving a claim, the AI analyzes eligibility for automated processing using predefined rules to assess coverage, liability, and other factors. It then takes appropriate actions, such as approving straightforward claims, rejecting those without coverage, or flagging complex cases for human review. [5]

Where AI Agents Will Be Adopted First

Due to several limitations hindering the widespread adoption of AI agents, only a few organizations are currently ready to embrace this technology. Enterprises are hesitant to integrate AI agents into their core operations, where unreliable decisions and actions could have detrimental consequences. Based on the capabilities and limitations of AI agents, their adoption is likely to follow these trends:

• • Tasks or businesses associated with low-risk impact from decision-making will adopt AI Agent quickly

Due to the early stage of AI agents, enterprises are cautiously adopting them in low-risk decision-making areas. Current real-world applications reflect this trend, focusing on controllable risk environments. E-commerce companies, for example, are leveraging AI agents for customer support and personalized recommendations, where the impact of decisions is less severe. However, for high-impact decision areas, human oversight remains crucial.Enterprises must carefully consider the implications of maintaining a human presence in these processes.

• • High data volume industries with robust data infrastructure will adopt AI agents more readily

Industries and organizations with robust digital data infrastructure and high data volumes are poised for faster AI agent adoption. These entities, like healthcare with its vast research data or customer service with its high volume of client interactions, already possess the digital foundation necessary for seamless AI integration. AI agents excel at processing and analyzing large datasets with speed and accuracy, surpassing human capabilities, making them invaluable in these data-rich environments. This advantage explains why tech companies and established enterprises, which typically manage data digitally, lead the way in AI agent adoption, while organizations relying on traditional, on-premise data storage may face a steeper learning curve.

• • Early AI Agent Adoption Will Occur Horizontally Before Vertically

AI agents are expected to be adopted across industries for broad, non-specialized tasks—such as customer service and productivity enhancement—before they gain traction in industry-specific applications. In healthcare, for example, AI agents could be used for scheduling, payment processing, and insurance claims, but they are not yet reliable for technical tasks requiring medical expertise. Similarly, legal firms may leverage AI agents for regulatory research or document summarization, but AI is not yet advanced enough to provide legal opinions.

Challenges Preventing Wide Adoption of Agentic AI

Despite Agentic AI seeming to provide promising benefits to enterprises, AI agents have not yet been adopted widely. Most current applications of AI agents focus on straightforward, low-risk impact tasks such as solving customer issues, automating document reviews, or sending emails and scheduling calls. The use cases of AI agents on more complex tasks have not been widely seen yet. Most of the ideal use cases described previously are in the experimental stage and rely heavily on human oversight, reducing the value of using an AI agent.

For AI agents to gain widespread adoption among enterprises for complex and high-stakes tasks, several challenges must be addressed. The key barriers preventing widespread adoption of AI agents include:

• • Lack of Trust in AI Agents Among Enterprises

A major obstacle to adoption is enterprises’ lack of confidence in AI agents. Organizations need transparency in how AI agents make decisions, process data, and ensure reliability. Startups developing AI agents must provide detailed documentation explaining how their models function and offer clear, understandable breakdowns for both technical and non-technical stakeholders. Additionally, security and compliance concerns are paramount. Enterprises need assurances that their data remains secure, and without strong security tools and frameworks, businesses will be hesitant to adopt AI agents.

• • Data Privacy and Security Concerns

Data privacy and security are paramount concerns for enterprises considering the adoption of agentic AI. As AI agents access vast amounts of sensitive data such as financial records and medical histories, these systems’ reliance on extensive data necessitates careful management of its acquisition, storage, use, and disclosure, along with strict adherence to relevant regulations and industry standards like SOC 2 Type I/II, ISO 27001, HIPAA, and GDPR.

• • High Upfront Costs and Integration Complexity

The high initial costs and complexity of integration remain significant hurdles. Data collection and training have yet to reach economies of scale that would drive costs down. Additionally, integrating AI agents into existing workflows requires extensive effort. For example, traditional hospitals looking to implement AI agents must migrate patient data from on-premise storage to cloud systems, establish policies for managing sensitive data access, and train staff to work effectively with AI technology.

Addressing these challenges is crucial for AI agents to gain widespread adoption across industries. As data privacy and security measurements are in place, trust issues are mitigated,and costs decrease, AI agents will become an integral part of enterprise workflow

The Emerging Business Opportunities in AI Agent Adoption

The challenges around AI agent adoption create business opportunities. As AI agents become more sophisticated and integrated into enterprises, various enabling technologies and services will be critical to accelerating their deployment. Startups and investors should monitor these key business opportunities closely, as these tools will play a pivotal role in facilitating AI agent adoption.

• • Model validation and monitoring tools that help increase trust and reliability

For enterprises to trust and effectively use AI agents, they must fully understand how AI systems make decisions rather than relying on them blindly. Startups in this space are developing software that functions as a quality control system for AI, enabling companies to monitor their AI models in real-time. These platforms provide transparency by explaining why AI makes specific decisions, reducing the “black box” effect and helping businesses enhance AI accuracy and reliability. Enterprises can view these platforms as dashboards that offer visibility into AI decision-making, ensuring fairness, accountability, and compliance with industry regulations.

• • Data privacy and security tools to ensure AI privacy and security[6]

To address data privacy and security concerns, enterprises should establish a comprehensive data governance framework.  However, policies alone are insufficient; effective implementation requires the right tools. The following tools can help organizations manage data privacy and maintain compliance with relevant regulations and industry standards.

• • • • Data masking and anonymization tools: Tools in this space help remove identity information from the data before using such data in AI models, mitigating the risks of unauthorized access and data breach.
      • Access control and audit tools: Tools in this group help manage data access control, allowing only authorized users to access data. Additionally, they could track data access, detect, and address unauthorized access to users. Once enterprises set up a data policy framework, they can deploy such frameworks by using these tools.
      • Data lineage and audit trail tools: These tools track the origin and movement of data, enabling users to understand its source, transformations, and ultimate destination. This creates an audit trail that demonstrates regulatory compliance.

• • AI Agent enablers that help bring the cost down and address integration complexity

Several key technological advancements are emerging to address the cost and complexity barriers associated with AI agent development and deployment. Among these are decentralized GPU cloud services and robust metadata management solutions.

• • • • Decentralized GPU Cloud Services

Training large language models and deploying AI agents in real-world applications requires significant GPU processing power, leading to high infrastructure costs. Decentralized GPU cloud services offer a solution by making computational resources more accessible and affordable. This model operates similarly to “Uber for the GPU world,” connecting users with idle GPU capacity to those who need affordable, scalable computing power. As AI agents continue to grow, decentralized GPU cloud services will likely gain traction, reducing dependency on traditional cloud providers and lowering entry barriers for AI startups.

• • • • Metadata Management

Metadata management is a foundational requirement for AI agents to function effectively and scale within organizations. It helps AI agents interpret context, maintain structured knowledge, and make informed decisions while facilitating agent communication, task orchestration, and knowledge sharing across distributed systems. Essential features of metadata management tools include data cleaning, classification, and organization to ensure accuracy, integrity, and consistency. This market is currently dominated by major tech players such as Databricks, Snowflake, and IBM, which provide robust solutions for enterprises looking to manage metadata at scale.

Conclusion

As AI agents continue to advance, the transition from their current capabilities to fully autonomous systems will take time. However, their potential to transform industries is increasingly evident. By boosting productivity, enhancing decision-making, streamlining workflows, and improving customer experience, AI agents are poised to play a pivotal role in the future of automation. Despite this promise, widespread adoption still faces significant challenges, including cognitive architecture development, enterprise trust, infrastructure readiness, and integration complexity. Nevertheless, businesses that rely on structured processes and large datasets are likely to be early adopters, paving the way for broader industry acceptance. As enabling technologies mature and organizations gain confidence in AI-driven decision-making, AI agents will gradually become an integral part of enterprise operations. The future of AI agents extends beyond mere automation—it lies in the development of intelligent, adaptive systems that work seamlessly alongside humans, unlocking new opportunities for innovation and efficiency across industries.

Author: Warittha Chalanonniwat (Paeng) 

Editors:  Krongkamol deLeon (Joy), Woraphot Kingkawkantong (Ping)

Reference

• https://medium.com/towards-data-science/ai-assistants-copilots-and-agents-in-data-analytics-whats-the-difference-2e63f8fb2384
• https://www.linkedin.com/pulse/when-choose-ai-co-pilot-vs-agentic-lata-tewari-r2mqc/
• https://beam.ai/agents
• https://www.miquido.com/blog/ai-agents-use-cases/
• https://bytebridge.medium.com/ai-agents-current-status-industry-impact-and-job-market-implications-f8f1ccd0e01f
• https://www.automaited.com/resources/blog/6-types-of-processes-that-are-suitable-for-rpa
• https://www.linkedin.com/pulse/when-choose-ai-co-pilot-vs-agentic-lata-tewari-r2mqc/
• https://medium.com/towards-data-science/ai-assistants-copilots-and-agents-in-data-analytics-whats-the-difference-2e63f8fb2384
• https://medium.com/@elisowski/ai-agents-agentic-ai-and-autonomous-ai-are-they-the-same-2ca7fbf5474a
• https://www.linkedin.com/pulse/agentic-ai-cybersecurity-autonomous-adaptive-proactive-protection-ucpqe/
• https://medium.com/@elisowski/ai-agents-vs-agentic-ai-whats-the-difference-and-why-does-it-matter-03159ee8c2b4
• https://www.teqfocus.com/blog/top-5-benefits-of-agentic-ai-driven-workflows-in-the-pharma-and-healthcare-industries
• https://botpress.com/blog/real-world-applications-of-ai-agents#:~:text=Virtual%20Assistants,personalized%20and%20effective%20over%20time.
• https://www.endava.com/glossary/agentic-ai#:~:text=Agentic%20AI%20evolved%20from%20conversational,language%20and%20autonomously%20make%20decisions.
• https://www.digitalocean.com/resources/articles/agentic-ai
• https://medium.com/spheronfdn/5-leading-decentralized-computing-platforms-transforming-access-to-gpu-computational-power-d9673fe4e40a
• https://www.madrona.com/the-rise-of-ai-agent-infrastructure/
• https://www.forbes.com/sites/robtoews/2024/07/09/agents-are-the-future-of-ai-where-are-the-startup-opportunities/
• https://www.grandviewresearch.com/industry-analysis/metadata-management-tools-market-report
• https://www.ibm.com/think/topics/explainable-ai#:~:text=Explainable%20AI%20is%20used%20to,putting%20AI%20models%20into%20production.
• https://www.nvp.com/blog/ai-agents-enterprise-bridging-gap-agentic-ai/
• https://www.forumvc.com/2024-the-rise-of-agentic-ai-in-the-enterprise
• https://medium.com/@mparekh/ai-long-road-ahead-for-ai-agents-for-enterprises-and-consumers-rtz-494-c86861de6c1d
• https://www.ankursnewsletter.com/p/horizontal-ai-agents-vs-vertical
• https://medium.com/@HitachiVentures/the-dawn-of-agentic-ai-transforming-enterprise-automation-and-the-future-of-work-e30844d41f1e
• https://www.linkedin.com/pulse/agentic-ai-here-its-shaking-up-business-landscape-tiarne-hawkins-yonff/
• https://www.cbinsights.com/reports/CB-Insights_Future-Workforce-AI-Agents.pdf
• https://www.everestgrp.com/automation/navigating-the-agentic-ai-tech-landscape-discovering-the-ideal-strategic-partner-the-rising-enterprise-adoption-of-agentic-ai-blog.html
• https://cloud.google.com/transform/101-real-world-generative-ai-use-cases-from-industry-leaders
• https://www2.deloitte.com/us/en/insights/industry/technology/technology-media-and-telecom-predictions/2025/autonomous-generative-ai-agents-still-under-development.html
• https://www.leewayhertz.com/ai-loan-underwriting/#loan-underwriting-processes-with-generative-AI
• https://blog.langchain.dev/what-is-an-agent/
• https://www.kenresearch.com/industry-reports/global-ai-agent-market
• https://www.sequoiacap.com/podcast/training-data-harrison-chase/?itm_medium=related-content&itm_source=sequoiacap.com#are-agents-the-next-big-thing
• https://www.pwc.com/m1/en/publications/documents/2024/agentic-ai-the-new-frontier-in-genai-an-executive-playbook.pdf
• https://www.ft.com/content/36785ec8-6f9f-455f-ac74-645bcaa9e221
• https://www.finextra.com/blogposting/27196/agentic-ai-the-competitive-edge-finance-and-insurance-companies-need-to-stay-ahead

[1] https://medium.com/@elisowski/ai-agents-agentic-ai-and-autonomous-ai-are-they-the-same-2ca7fbf5474a

[2] https://medium.com/@elisowski/ai-agents-vs-agentic-ai-whats-the-difference-and-why-does-it-matter-03159ee8c2b4

[3] https://davoy.tech/agentic-ai-capabilities-and-applications/

[4] https://www.linkedin.com/pulse/agentic-ai-healthcare-real-world-use-cases-revolutionizing-hgobe

[5] https://beam.ai/resources/case-studies/dutch-insurance-claims-processing

[6][6] https://www.zartis.com/ai-and-data-protection/how-to-protect-your-ip-while-using-ai/

As temperatures rise and climate change has an increasingly severe impact on human life, individuals and institutions alike are becoming more concerned regarding greenhouse gas emissions and the adoption of technologies which can decarbonize the world.  While the majority of greenhouse gas emissions is generated by industrial and commercial activity^[1] and the decarbonization of one individual’s life will have minimal impact on the world’s carbon footprint, it is still important to consider the decarbonization of consumer life.  Not only is there overlap between the technologies that can decarbonize consumer life and commercial activities, but corporations serve the end consumer – shifting consumer demand away from high emissions goods and services will incentivize corporations to boost supply of sustainable technology.  Increased consumer decarbonization may lead to a positive feedback loop with consumers placing more pressure on corporations to decarbonize and governments to drive policies supporting environmental sustainability.

Where should investors look to maximize the impact of their dollars on decarbonizing consumer life?  What will drive consumers to opt for sustainable solutions, and how are ecosystem players such as startups, governments, and financial institutions working to accelerate consumer adoption of green technology?  This article will explore the state of the world and the potential solutions for speeding up the voluntary decarbonization of consumer life.

State of the World: Consumer Emissions

Research shows that the major sources of emissions in consumer life are Transportation and Housing, followed by Food as a relatively distant third major source of emissions.^[2]

Total carbon footprint of the typical U.S. household: 48 t CO₂e/yr.  Blue indicates direct emissions, green indicates indirect emissions.  Source: “Quantifying Carbon Footprint Reduction Opportunities for U.S. Households and Communities”, 2010

As shown in the figure above, the single largest source of direct consumer emissions is motor vehicle fuel.  While this figure may vary in different parts of the world (particularly those with a less dominant car culture than the US), it is apparent that the most impactful way to decarbonize consumer life is to minimize the use of traditional ICE vehicles, either through increased adoption of public transportation or adoption of electric vehicles.

Consumers generate emissions from Housing in the form of electricity usage; as electricity grids in most countries are still driven by fossil fuels, this generates indirect emissions in consumer life.  Removing emissions from this sector therefore means reducing energy consumption from the grid via residential solar installation.  Further, reducing total energy consumption by the consumer will automatically reduce consumption from the grid.  The primary usage of electricity in residential buildings is related to Heating, Ventilation, and Air Conditioning (HVAC) systems for heating and cooling buildings.  Therefore, consumers can further reduce emissions by installing energy efficiency tools or optimizing home insulation (allowing HVAC systems to operate more efficiently).

While the data above is focused on the US market, there are similar findings throughout the rest of the world.  Studies in the European market also look at housing, mobility and food as the key emissions sectors for policy makers to tackle.  The importance of energy and housing in consumer emissions is even more prevalent in developing markets and low-income households as the bulk of these emissions are driven by energy consumption.^[3]  Despite the availability of green alternatives for transportation and housing, adoption by consumers has been limited.  EVs only accounted for 12% of US passenger vehicle sales in 2023^[4]; only 5% of US homes have rooftop solar installed.^[5]  In recent years, the rate of adoption has improved; the next sections will discuss the driving factors that can contribute to increased adoption as well as various approaches by governments, startups, and financial institutions which will continue to accelerate consumer decarbonization.

How to Accelerate Consumer Adoption of Sustainable Technology

Given the emissions sources identified above, what is the key to decarbonizing consumer life?  While carbon credits are often cited as a market mechanism for decarbonization, the uncertainty surrounding their value and effectiveness makes carbon credits a poor option for driving consumer decarbonization.  On the individual scale, carbon credits provide insufficient financial incentive (if any), and alternatives such as green reward tokens have limited use and may not be particularly useful to the average consumer.  Unlike the corporate sector, there are also no regulations enforcing consumers to decarbonize their lives, thus all consumer decarbonization is voluntary.

For consumer transportation and housing, much of the technology required for decarbonization is already available and proven to work, but adoption of these technologies has been slow to materialize.  Despite multiple reports indicating the majority of consumers are concerned about sustainability and are looking to adopt sustainable technology and products, concerns around additional costs of sustainable living may prevent rapid adoption.^[6]  As will be discussed below, it is also often the case that choosing the sustainable option will not result in higher cost of living.  In these instances, if consumers are still slow to make the choice to decarbonize, it suggests that there are still non-financial barriers to adoption that must be overcome.  In considering why consumers have been slow to adopt sustainable technology and how governments and businesses can improve adoption, it is useful to think about the triggers for changes in human behavior (such as internal motivations, financial considerations, or simply removal of friction).  This consideration leads to the assumption that voluntary decarbonization by consumers will require at least one of the following conditions: a shift in consumer mindset, a financial incentive, and/or sufficient quality and access to sustainable alternatives.

The required shift in consumer mindset is for consumers to see sustainability as necessary to the point that they are willing to make sacrifices or pay a green premium to adopt sustainable technologies.  By definition, this is a willingness to adopt new technology prior to price parity.  This kind of radical shift in mindset is difficult to control or implement, and likely will take a long time (and possibly an extreme and harmful crisis to trigger a true shift in mindset).  Further, trying to push consumers to pay a premium for green technologies also presents problems with respect to income inequality – many consumers in developing markets do not have the financial resources to pay any kind of premium.

Financial incentives in this framework refer to any mechanism that makes the adoption of sustainable technology cheaper than the traditional alternative.  This includes direct subsidies or tax incentives, as well as financing models that can alter payment horizons.  In the context of this framework, “financial incentives” also includes technological improvements which can decrease the cost of production of green goods or services below their non-sustainable counterparts.

Sufficient quality and access to sustainable alternatives may be hard to quantify, but for the purpose of this framework, “sufficient” means enough to enable consumers to willingly adopt technology to decarbonize their lives when the “green” and “not green” options are equal in price.  The assumption that consumers will adopt at price parity (whether that parity is achieved by technological/production improvements to decrease the intrinsic cost or through the aforementioned financial incentives) ignores the non-monetary and unquantifiable aspects of a product or technology.  For example, switching to public transportation instead of driving is likely to save a consumer money (and certainly is unlikely to require them to pay a green premium), however in many cases switching to public transportation may increase the difficulty for an individual to reach their intended destination.  Alternatively, green solutions may be equivalent in price, but consumers may lack awareness of how to access the relevant subsidies or lack understanding of what solutions to implement (for example, what products improve home insulation and therefore lead to lower HVAC energy usage).  “Sufficient quality and access to sustainable alternatives” is therefore meant to improvements in technology or business models which can reduce these non-financial barriers to adoption.

While a shift in consumer mindset may be necessary in the long run to drive full decarbonization of consumer life, it is clear that in the short run, ecosystem players hoping to drive consumer decarbonization should be focusing on how to improve financial incentives, quality and access to decarbonization technologies, or both.  This can be achieved by reducing the financial barriers through government subsidies or innovative financing models, or by improving the access to or useability of green technology.

How Are Ecosystem Players Tackling Consumer Decarbonization?

Using the framework above, effective models for boosting consumer adoption of sustainable technology can be divided into two categories: models for providing financial incentives or reducing the financial barriers to adoption, and businesses aimed at improving the access to and useability of decarbonization technology.  Initiatives to support consumer decarbonization can come from across the ecosystem. It may appear that governments and financial institutions are best placed to focus on financial incentives while startups and technology companies focus on improving access and useability, however, as will be demonstrated below, there are opportunities for all of these players in both categories.  It is also important to note that financial incentives alone may not be enough – the most effective models often combine financial incentives with frictionless access/useability.

Financial Incentive Models

Governments are typically the best (if not only) entity for providing direct financial incentives.  Decarbonization creates a public good, and if there is an increased cost for individuals to adopt sustainable technology, governments are the right entity to cover any increased costs to encourage faster adoption by individuals.  There are many examples around the world of governments stepping in with policy initiatives designed to financially incentivize consumers to look for sustainable options.  The US Inflation Reduction Act provides a federal tax credit for 30% of the cost of residential solar installation.^[7]  In Thailand, the government has allocated 7.12 billion THB to fund its EV subsidy program.^[8]  The EV 3.5 scheme provides consumer subsidies for EV car purchases of up to 100,000 THB (depending on the vehicle’s battery size and the year of purchase), as well as decreasing excise taxes (from 8% down to 2%) and import duties (by up to 40%) in order to make EV adoption more financially attractive for consumers.^[9]

While subsidies are the simplest form of financial incentives to understand, efforts to financially incentivize voluntary consumer decarbonization are not limited to governments; financial institutions and technology startups are also finding new financing models to encourage consumers to adopt sustainability.  Banks and solar installers can collaborate to provide consumers with different financing options such as solar leases or power purchase agreements (PPA) to decrease the upfront costs of solar installation.  In Southeast Asia, technology startups like Helios and Okapi work with partners across the residential solar and financial supply chain to enable homebuyers to access affordable solar installation.  On the other end of the cost-benefit spectrum, companies like Powerledger are leveraging blockchain technology to enable P2P energy trading not only for businesses, but within residential communities.^[10]  At scale, P2P energy trading platforms could enable consumers to generate additional future income from their solar panels, providing more attractive returns in exchange for the upfront cost of adopting solar technology in their homes.

Models to Improve Access and Useability

The other significant opportunity to decarbonize consumer life lies in improving consumer access to sustainable technology and improvements in the practical useability of sustainable technology.  This factor is often the last remaining gap to cross to ensure frictionless adoption for consumers – if adoption of a technology is falling short of targets despite financial incentives, it is likely that there is still a quality gap, an implied sacrifice consumers are being asked to make in their daily lives to shift to sustainable technology.  For example, one major concern for consumers when considering purchasing an electric vehicle is the charging infrastructure and the distances they may need to travel in between available charge points.  Governments, financial institutions, and investors should consider that investing in EV infrastructure is essential to accelerating the switch from traditional ICE vehicles to electric alternatives.  In Thailand, Kasikornbank has launched WATT’S UP, an EV motorcycle rental and battery swapping platform to promote easier access and increased usage of EV motorcycles.^[11]  Several startups throughout the Southeast Asian region are working to enable cross-platform charging and ensure drivers have easy access to multiple charging options in the same way that ICE vehicles are able to refuel at any gas station regardless of vehicle brand.  Alternatively, startups are also tackling green innovations in the public transportation sector, which could make the switch to public transportation more feasible for consumers.  Muvmi, a Thai startup, provides electrified last-mile transport to bridge the gap between established public transit stations and the consumer’s final destination (such as office buildings, popular shopping destinations, or homes).

When looking at ways to decarbonize residential housing, electrification can have a major impact on a consumer’s carbon footprint.  Home electrification projects such as installation of solar panels or electric heat pumps, deployment of EV chargers and battery storage, or weatherization to reduce in-home energy consumption can both decarbonize the home and lead to long-term cost savings.  Startups across the globe, but particularly in developed markets like the US, are leveraging digital technology to solve these kinds of problems.  Companies like Pika and Zero Homes have designed software solutions to enable contractors to simplify the sales process for home electrification projects and improve sales efficiency to speed up adoption by consumers; many startups combine digital tools for sales agents and installers with streamlined payment processes to automate the consumer’s access to financial incentives and minimize the friction in the homeowner’s decision making process.

Conclusion

While many discussions on consumer decarbonization may naturally focus on food or consumer goods (which are often more visible issues in the media), a greater degree of impact can be created by focusing solely on decarbonizing home energy usage and consumer transportation.  Past efforts to push decarbonization have also relied on consumers to make the “right” choice, requiring them to pay green premiums or to make sacrifices in their daily lives in order to adopt sustainable technology, which largely results in slow voluntary adoption of said technology.  Consumer mindsets may slowly shift over time, but there is an opportunity in the current market to speed up adoption in the near-term by focusing on reducing the financial barriers and improving accessibility and useability of green solutions.  Governments seeking to hit Paris Agreement targets and financial institutions looking to achieve their net-zero commitments are actively providing financial incentives and new payment models which can reduce the financial burden for consumers to decarbonize.  As discussed previously, financial incentives alone may not be sufficient to trigger mass adoption of green technologies, and therefore many businesses seeking to accelerate their sale of such technologies to the consumer segment look for opportunities, investments, and business models that make the green options easier to use, easier to access, and requiring minimal change from the consumer’s perceived status-quo.  By focusing on these specific areas, ecosystem players seeking to accelerate consumer decarbonization can identify opportunities which can maximize the impact of their investment.

Author: Krongkamol deLeon (Joy)

Editors: Benjamas Tusakul (Air), Woraphot Kingkawkantong (Ping)

Reference

[1] https://cz.boell.org/en/2023/07/26/individual-carbon-footprint-how-much-does-it-actually-matter

[2] Christopher M Jones and Daniel M Kammen, “Quantifying Carbon Footprint Reduction Opportunities for U.S. Households and Communities,” Environmental Science & Technology, Vol 45/Issue 9,  https://pubs.acs.org/doi/10.1021/es102221h

[3] https://link.springer.com/article/10.1007/s10018-019-00253-7?fromPaywallRec=false

[4] https://www.rabobank.com/knowledge/d011429876-the-rise-of-electric-vehicles-in-the-us-and-the-road-ahead

[5] https://www.rewiringamerica.org/research/pace-of-progress-home-electrification-transition

[6] https://www.bain.com/insights/ten-takeaways-from-our-2024-sustainability-survey-of-consumers-infographic-ceo-sustainability-guide-2024/

[7] https://www.energy.gov/eere/solar/homeowners-guide-federal-tax-credit-solar-photovoltaics

[8] https://www.bangkokpost.com/business/motoring/2871632/cabinet-allots-b7-12bn-for-ev-subsidies#:~:text=Under%20the%20EV%203.5%20scheme,to%2050%2C000%20baht%20per%20vehicle.

[9] https://www.ryt9.com/s/iq03/3479762

[10] https://powerledger.io/platform-features/xgrid/

[11] https://www.kasikornbank.com/en/news/pages/wattsup.aspx

Cybersecurity refers to the practice of protecting computer systems, networks, and data from unauthorized access, theft, damage, disruption, or other forms of attack. The practice in cybersecurity can involve several technologies, and Artificial Intelligence (AI) is one of the key enabling technologies among many.

This article will explore the importance and use cases of AI in cybersecurity, which bring about both benefits and risks, some risk mitigation methods that are currently being discussed, including how to balance human involvement with AI operations in cybersecurity, implications of AI in cybersecurity for financial service industry, and future trends of AI development in the cybersecurity landscape. This article, inspired by the author’s interest in exploring the role of AI in cybersecurity and finding the right balance with human involvement, aims to share insights from research, the author’s thought processes, and key findings from the exploration.

What is the Role of AI in Cybersecurity? Why is it Becoming Increasingly Important Now?

AI can be categorized into predictive AI and generative AI based on functions they perform. Predictive AI involves models trained on large datasets to identify patterns, correlations, and trends, often used for tasks such as forecasting, classification, and risk assessment. Generative AI, on the other hand, focuses more on generating new content such as text, audio, video, etc., and understand unstructured data.

Predictive AI has been used in cybersecurity since the late 1980s to detect abnormal activities, recognized for its speed and accuracy beyond human capability. However, recent advances in generative AI and machine learning (ML) have contributed both positively and negatively to the cybersecurity realm. To cyber criminals, generative AI, together with ML, allows them to launch more sophisticated attacks at a larger scale such as crafting more realistic phishing emails or creating malware with the ability to adapt its own code to avoid detection by traditional security system. On the other hand, generative AI can be used by cybersecurity team as a countermeasure to enhance threat intelligence by automating learning of unstructured threat data and build a new capability to analyze more qualitative data to improve threat detection.

AI’s capabilities in cybersecurity benefit all organizations, but for large corporations with ample resources, AI improves efficiency and speeds up responses to threats. For SMBs with limited budgets, the benefits are more accentuated in terms of cost reduction for hiring cybersecurity personnel.

With its promising capabilities, organizations are increasingly adopting predictive AI, generative AI, and ML for their cybersecurity practices to boost efficiency and reduce costs by automating labor-intensive tasks. This is, in one way, evidenced through a survey of 800+ senior management by Arctic Wolf where 98% of respondents plan to allocate some portion of their upcoming cybersecurity budget towards AI. Within those, 52% are dedicating over a quarter of their budget in this area.

Note: The term AI used in this article onward means both predictive AI and generative AI together with the capabilities of ML.

AI Capabilities and its Implications to Cybersecurity

AI benefits and capabilities for cybersecurity can be explained through 6 core functions of cybersecurity, which will be further elaborated in the next paragraph. While the benefits offered are not negligible, there are also some restrictions and risks associated with adoption of AI for cybersecurity practices. Organizations must be prepared to mitigate such risks to optimize AI performance.

Cybersecurity can be divided into 6 functions according to the NIST (National Institute of Standards and Technology) framework, a world-renowned framework adopted by global organizations such as Saudi Aramco, Israel National Cyber Directorate, University of Chicago, etc. The definition of each according to NIST is as shown below.

Figure 1: NIST Cybersecurity Framework

1. GOVERN addresses an understanding of organizational context; the establishment of cybersecurity strategy and cybersecurity supply chain risk management; roles, responsibilities, and authorities; policy; and the oversight of cybersecurity strategy.
2. IDENTIFY understands the organization in deeper detail (e.g. data, hardware, software, systems, facilities, services, people) and related cybersecurity risks, which enables an organization to prioritize its efforts consistent with its strategy and direction as identified under GOVERN.
3. PROTECT supports the ability to secure organizational assets to prevent or lower the likelihood and impact of adverse cybersecurity events, as well as to increase the likelihood and impact of taking advantage of opportunities.
4. DETECT enables the timely discovery and analysis of anomalies, indicators of compromise, and other potentially adverse events that may indicate that cybersecurity attacks and incidents are occurring.
5. RESPOND supports the ability to contain the effects of cybersecurity incidents.
6. RECOVER supports the timely restoration of normal operations to reduce the effects of cybersecurity incidents and enable appropriate communication during recovery efforts.

For more information or examples of activities under each function, please visit NIST website.

To understand which tasks are suitable for AI in each function, it is essential to first analyze the expected outcome of functions, which can be done by understanding the objectives. Then, the key success factors can be determined based on qualities that will lead to better outcome of the function, which in turn will be mapped with AI’s unique capabilities to derive suitable tasks under each function. Please see below an analysis of the expected outcome and key success factors for the 6 functions, together with AI/ML capabilities.

Figure 2: Analysis of AI/ML capabilities for Cybersecurity Core functions

Given the unique strengths of AI, particularly its capability to handle vast amount of data, automate routine tasks, and perform real-time actions, AI can help enhance an organization’s cybersecurity posture through all 6 activities, albeit at different capacity.

Even with all the promising capabilities that AI provide, it is a double-edged sword. There are still some downsides that users need to be aware of. There are two major concerns commonly mentioned and discussed across sources.

1. Ethical concerns:
   • Bias and discrimination in decision-making: This can stem from non-diverse training data set or bias from machine learning process of non-relevant input factors such as gender, race, etc. In cybersecurity, for example, AI might flag certain groups more frequently, leading to unequal treatment. This is especially concerning when using AI in areas like fraud detection or risk assessments where fairness is crucial.
   • Privacy concern from data used to train AI: Data used to train AI is usually retrieved from production databases which might contain personal sensitive information, leading to concerns about data leakage and personal data rights. Data used to train AI for cybersecurity, specifically, often includes personally identifiable information such as user behaviors and biometric information. Strong security measures must be implemented end-to-end from the origination and transmission of data to the handling of data after its use, to ensure no data leakage from the additional exposure.
   • Lack of transparency and explainability: AI is often regarded as a Black Box system, where users can only see inputs and outputs, but not the processes in between. This lack of explainability can cause trust issues, as cybersecurity teams might not fully comprehend why an AI system flags certain behaviors as malicious or overlooks potential threats. Transparency is key to ensuring that AI’s actions align with the organization’s cybersecurity objectives.
2. Potential Mistakes from AI:
   • Mistakes happen from a model itself such as:
     • A generative AI hallucination – AI models generate incorrect or misleading results, which are caused by a variety of factors, including insufficient training data, incorrect assumptions, and etc. according to Google Cloud.
     • An overfitting of models – AI algorithm fits too closely or even exactly to its training data, resulting in a model that can’t make accurate predictions or conclusions from any data other than the training data according to IBM.
     • An example of this type of mistake in cybersecurity regime is when an AI model flags a legitimate activity as malicious (false positives) or fails to detect actual threats (false negatives).

• • Mistakes happen from malicious actions such as:
    • Data manipulation – Cyber threat actors manipulate data consumed by AI algorithms. By inserting incorrect information into legitimate but compromised sources, they can “poison” AI systems, causing them to error out or export bad information, according to BlueVoyant. For example, attackers might alter logs or feed deceptive data into AI-driven monitoring systems to avoid detection. When AI gradually recognizes such pattern as normal, attackers can then utilize this attack vector for an actual offense.
    • Model theft – AI model itself is compromised and reversed engineered by attackers to find vulnerabilities of the model. Attackers can then exploit weaknesses discovered to launch undetected attacks.

There are currently two main approaches to mitigate these concerns from organizations’ side which are the use of technological solutions and human involvement and supervision. These two options are generally utilized in combination to tackle the concerns.

1. Technological solutions are seen more often to solve the following concerns.
   • Privacy concern – This concern can be mitigated by using tools to generate synthetic data or perform data masking. There are multiple players providing these solutions such as betterdata, Hazy, Mostly AI.
   • Lack of transparency and explainability – One way to address this concern is to use AI solutions with clear documentation on decision-making processes which can be audited and customizable as needed.
   • Potential mistakes from both a model itself and malicious actions – There has been discussion about building an AI agent to work for humans. AI agent, according to IBM, is “a system or program that is capable of autonomously performing tasks on behalf of a user or another system by designing its workflow and utilizing available tools”. In this case, AI agent can be specifically trained to validate the output of another AI against expected outcome to prevent mistakes.
2. Human involvement and supervision
   • Bias and discrimination in decision making – To prevent bias and discrimination in AI, ethicists should be integrated into AI development and deployment teams to ensure that training datasets are unbiased, and outputs are rigorously tested for fairness. Geoffrey Hinton emphasized this approach during a session at Collision 2024. More information on AI ethicists can be found here.
   • Potential mistakes from both a model itself and malicious actions – Humans can serve as validators of AI outputs, identifying and correcting any mistakes made by the AI model. This approach follows the ‘Maker-Checker’ principle, ensuring an additional layer of oversight and accountability.

Balancing Human Involvement in the Age of AI Cyber Defense

It is undeniable that AI has unlocked the level of efficiency that is not previously achievable by humans. On the other hand, a human component in cybersecurity operations is still mandatory to mitigate ethical concerns and reduce risks from AI mistakes. Thus, balancing the usage of both components is key to resilient cybersecurity.

Achieving an optimal balance between humans and AI requires a clear understanding of their respective strengths and limitations. Cybersecurity tasks can be categorized into two key areas: operational capabilities and intelligent capabilities. Operational capabilities ensure tasks are executed effectively and efficiently, while intelligent capabilities ensure tasks are carried out responsibly and aligned with an organization’s goals. This analysis helps determine which cybersecurity activities are best suited for AI and which should remain under human supervision. The table below highlights the strengths and limitations of both AI and humans in these areas, with green cells indicating strengths and red cells indicating limitations.

As seen in the table above, human and AI possess different strong traits, suggesting rooms to efficiently divide tasks between the two. While small, the overlapping greens or reds point to possibilities of collaboration. When mapped with the NIST framework for cybersecurity, leading roles for each of the 6 tasks and activities that each can perform to strengthen the security postures are as identified in the following figure, where human is tasked to lead roles that require high level of strategic decision-making and communication, while AI is expected to lead the execution according to the preset policies which are more operational by nature.

Figure 3: Human and AI Collaboration on the Cybersecurity’s Core Functions

AI in Cybersecurity for Financial Service Industry

Financial services are frequently cited as a top target for cyber-attacks, with the industry incurring the second-highest breach costs, averaging nearly $6 million annually, according to Nvidia. Due to the high volume and value of monetary transactions, financial institutions are particularly vulnerable to identity fraud and transaction fraud. These forms of fraud are highlighted because identity theft can give attackers access to accounts allowing them to perform fraudulent transactions, while transaction fraud directly compromises the transfer of funds, making them critical concerns for the sector.

Preventions for these two types of frauds often require analysis of high volume of data and involve routine operations in PROTECT, DETECT, and RESPOND functions, in which AI excels at leading the tasks. Therefore, AI is undeniably an effective and efficient defense tools against these frauds for financial service providers to manage the risks. This section will deep dive into how AI has helped the financial service industry mitigate risks of these two frauds and some solution providers.

Identity fraud is “the crime of using someone’s personal information in order to pretend to be them and to get money or goods in their name” according to the Cambridge Dictionary. Some examples of prominent identity frauds include:

1) Phishing – malicious actor sends a phishing content through channels such as email, text message, to account owners, luring them to provide personal credentials or financial information;

2) Fake website – threat actor creates a fake website, looking like legitimate and trustworthy one, deceiving account owners to input financial information or make false financial transactions; and

3) Data breaches – cybercriminal gains access to account owner credentials and information through unauthorized database access or other forms of records.

Transaction fraud is “any deceptive activity intended to acquire money, goods or services during a financial transaction” according to Datavisor. Transaction fraud typically happens after identity fraud, if not at the same time. Cybercriminals use credentials received from identity fraud to perform financial transactions such as using credit card information for unauthorized purchases, using login credentials to perform money transfer to their own accounts.

Several large banks around the world have integrated AI into their cybersecurity measures to protect their customers and minimize their financial losses and reputation damages. Many have announced their strategies on using to address cybersecurity challenges including Bank of America, JPMorgan Chase, KBank, BNP Paribas, Mitsubishi UFG and more. Some outstanding use cases of AI as countermeasures for these frauds being implemented in the financial service sector are shown below.

Conclusion

AI has become essential in cybersecurity, offering new capabilities for both attackers and defenders. While AI enables faster, more widespread cyberattacks, it also empowers defense mechanisms to counter threats at unprecedented speed and scale. The effectiveness of AI grows with more data, making it a race where “data is the new oil,” as Clive Humby noted. However, AI is a double-edged sword, with ethical concerns and potential errors posing significant risks. To mitigate these, the industry is balancing AI’s strengths, like rapid data analysis and automated responses, with human oversight for tasks requiring context and nuanced judgment. One of the prominent use cases is for the financial service industry, which deals with high volume and value of monetary transactions. AI is being widely adopted to prevent identity fraud and transaction fraud due to its strengths in speedy high volume data analysis and routine task automation.

The AI era is just beginning, with many future possibilities to strengthen cybersecurity. One promising initiative is cross-environment intelligence, where AI models can learn from data across multiple organizations without exposing sensitive information, creating real-time collective intelligence. However, this requires central coordination and standardized integration across systems, making it a work-in-progress. Another development is the rise of AI agents, which can integrate with systems to automatically perform cybersecurity tasks using available tools and applications, and collaborate with each other, like humans, to enhance security and push automation further in cybersecurity operations.

As we venture into this ever-evolving landscape of cyber threats, organizations must stay informed on emerging trends and technologies to remain resilient, with AI being at the forefront. However, the use of AI in cybersecurity will require human supervision to ensure ethical outcomes, prevent mistakes, monitor undocumented data, and make strategic decisions. Only with this balance between AI and human oversight can organizations fully harness the potential of AI to effectively enhance their cybersecurity defenses.

Author: Benjamas Tusakul (Air)

Editors: Wanwares Boonkong (Pin), Woraphot Kingkawkantong (Ping)

Reference

https://www.sophos.com/en-us/cybersecurity-explained/ai-in-cybersecurity

https://www.engati.com/blog/ai-in-cybersecurity

https://talkbusiness.net/2024/03/the-pros-and-cons-of-ai-in-cyber-security/#:~:text=The%20Cons%20of%20AI%20in%20Cyber%20Security&text=AI%20tools%20themselves%20have%20become,Deepfakes%20are%20also%20a%20risk.

https://www.securitymagazine.com/articles/99487-assessing-the-pros-and-cons-of-ai-for-cybersecurity

https://www.statista.com/statistics/1382266/cyber-attacks-worldwide-by-type/

https://www.weforum.org/agenda/2024/01/cybersecurity-cybercrime-system-safety/

https://www.entrepreneur.com/science-technology/how-ai-can-improve-cybersecurity-for-businesses-of-all-sizes/476727#:~:text=Artificial%20intelligence%20plays%20a%20dual%20role%20in%20cybersecurity,growth%20of%20cybercrime%20in%20the%20next%20few%20years

https://www.ey.com/en_gl/insights/consulting/transform-cybersecurity-to-accelerate-value-from-ai

https://arcticwolf.com/resource/aw/the-human-ai-partnership 

https://academia.co.uk/ai-versus-human-collaboration-for-a-secure-digital-future/

https://secureframe.com/blog/ai-in-cybersecurity

https://www.paloaltonetworks.com/cyberpedia/generative-ai-in-cybersecurity

https://outshift.cisco.com/blog/adopting-ai-security-operations

https://www.techmagic.co/blog/ai-in-cybersecurity/

https://www.americanbanker.com/news/can-ai-help-when-a-scam-is-invisible-to-the-bank

https://innov8tif.com/6-ways-ai-is-fighting-back-against/

https://www.cio.com/article/190888/5-famous-analytics-and-ai-disasters.html

https://www.splunk.com/en_us/form/state-of-security.html

https://www.tableau.com/data-insights/ai/advantages-disadvantages

https://www.theimpulsedigital.com/blog/ai-vs-human-intelligence-exploring-the-advantages-and-limitations/

https://www.radware.com/blog/security/threat-intelligence/2024/06/beyond-chatgpt-how-ai-agents-are-shaping-the-future-of-cyber-defense-and-offense/

Image generated by Gemini

Blockchain technology has come a long way since its emergence with Bitcoin (Blockchain 1.0) and the introduction of Ethereum and smart contracts. Core promises including increased decentralization, transparency, traceability, efficiency, and automation have fueled its development. Today, there’s a tremendous push to build the infrastructure of the future, with a vision to disrupt and create a better financial landscape. With growing clarity on blockchain’s potential and evolving regulations, the industry sets its sight on mass adoption. Today, the focus on “real-world use cases” becomes topics of conversation. These use cases span from cross-border payments and asset tokenization for democratized investment to identity verification.

However, a critical first step is taking a step back and addressing fundamental questions: Why focus on the Reals (Real Money, Real Assets, Real Identity)? What problems are we solving? Why is now the time?

This article dives into defining the Reals, answering these crucial questions related to real-world use cases. We will also explore the why and how Financial Institutions (FIs) can participate in driving mainstream adoption and identify the gaps they need to address to pave the way for the future of finance.

Overview of the “Reals”

Before diving into the core questions, let’s establish a clear definition and framework of “the Reals”.

For this article, we will utilize the definitions established by Quarix, a blockchain infrastructure developed by Orbix Technology under Kasikornbank. Their framework resonates well with our focus, offering comprehensive definitions in the financial context while emphasizing the crucial element of “real” – a secure and verifiable representation within the digital realm.

The following is the breakdown of the “Reals”:

• Real Money: Tangible money based on national currency that can be used for all practical purposes
• Real Assets: Tokenization of real-world assets such as bonds, capital market assets etc. on the blockchain
• Real Identity: Secure on-chain verification of users to ensure every user’s authenticity and identity

By focusing on these Reals, we can explore how blockchain technology is revolutionizing the way we manage and interact with money, assets, and identities in the digital age.

Why ‘Reals’ Matter

The widespread adoption of blockchain technology hinges on its ability to tackle real-world problems, particularly within the financial sector that has numerous inefficiencies and vulnerabilities and demand better and more streamlined products and services. Blockchain’s core strengths – decentralization, immutability, and traceability – offer immense potential for streamlining operations across various industries. However, focusing on applications anchored in the Reals framework (Real Money, Real Assets, and Real Identity) unlocks its true potential for financial services.

There are three main fundamental questions that we need to answer related to the development of real-world use cases leveraging blockchain technology. Let’s deep dive into each of the questions.

1. Why Do We Need to Focus on the “Reals”?

The financial landscape is ripe for disruption. The “Reals” framework can address fundamental challenges hindering the current financial landscape and unlock the true potential of blockchain technology. Here’s why focusing on Reals is crucial:

• • Grounding Innovation in Reality: The “real” in Reals emphasizes that these applications are grounded in verifiable reality, not purely speculative ventures. This focus fosters trust and confidence in blockchain’s ability to revolutionize financial services.
  • Solving Existing Problems: Traditional systems managing money, assets, and identities are often plagued by inefficiencies and security vulnerabilities. Blockchain’s cryptographic technology safeguards data integrity, making it incredibly difficult to hack or manipulate information. Additionally, blockchain technology has the potential to build new business models that could generate new revenue streams and optimize costs.

• • Achieving Mainstream Adoption: By focusing on the Reals, blockchain addresses issues directly relevant to everyday users and businesses. This focus on tangible benefits creates a compelling case for adoption and drives user confidence in the technology.

Additionally, the “Reals” framework offers a transformative approach to blockchain, fostering trust and revolutionizing finance for all stakeholders. It empowers the blockchain community to drive mass adoption by focusing on real-world problems and fostering user trust. For financial institutions, Reals unlock a competitive edge with innovative products and services and cost optimization, while enhancing inclusivity. Regulators benefit from a more robust and efficient financial system through transparency, traceability, and secure identity management. By prioritizing tangible applications grounded in reality, Reals pave the way for a trusted future of finance.

2. What Problems Are We Trying to Solve?

While the possibility for fantastical application exists, grounding blockchain in the “real-world” in the financial sector addresses several key issues:

• • Inefficiencies within Traditional Finance Realm: Traditional systems for managing money, assets, and identities can be complex and slow, leading to inefficiencies and errors. Blockchain offers automation capabilities such as document verification and processing that can streamline the tracking of goods and documents. As automation occurs across a distributed network, it eliminates a single points of failure and ensure transparency in every step. For instance, a World Bank report highlights that trade finance inefficiencies cost an estimated $1.5 trillion globally each year. By streamlining trade finance processes with tokenized assets, the technology can help unleash the potential for cost savings.
  • Vulnerable Security: The digital world is rife with opportunities for manipulation, fraud, and data breaches that expose sensitive financial information and user identities. Traditional systems are vulnerable to these threats. Blockchain technology offers a solution: its tamper-proof architecture creates a verifiable audit trail, significantly reducing fraud risks. Additionally, blockchain’s cryptographic protocols make it incredibly difficult to hack. Data from Chainalysis, reported by The Record, highlights this issue – DeFi platforms alone saw $1.1 billion stolen in 2023, with incidents rising from 219 in 2022 to 231. By focusing on the Reals, blockchain can create a more secure ecosystem for financial transactions and identity management.
  • Lack of Transparency: One of blockchain’s core strengths is transparency. All transactions are recorded on a shared ledger, accessible to authorized participants. This shared ledger acts as a single source of truth, providing real-time visibility into the status of transactions and movement of goods for all parties involved. Walmart Canada exemplifies this by leveraging blockchain technology to tackle supple chain challenges and develop a solution for invoices management. With clear audit trail that blockchain technology enables, it fosters another level of trust in the new financial system. Unlike traditional opaque systems, everyone involved can track the progress and verify the legitimacy of transactions.

3. Why Do We Need to Focus on It Now?

The urgency to address the real-world problems with blockchain is intensifying mainly due to several converging factors:

• • Unlocking Sustainable Growth: Financial activities are rapidly moving online driven by higher efficiency and better convenience. This increasing reliance on digital systems emphasizes the critical need for secure and trustworthy infrastructure. With its immutability and transparency, blockchain technology is well-positioned to address the need, build trust and enabling sustainable growth in the new financial landscape.    

• • Meeting Evolving Customer Demand: In a digital world demanding ever-better financial solutions, businesses and consumers prioritize efficiency, security, and cost-savings. Traditional systems often struggle to meet these expectations. Blockchain technology offers a compelling solution, delivering substantial improvements at minimal cost. Customers are unconcerned with the technology itself, but rather the improved product or service. Blockchain excels in this matter, streamlining back-end processes with smart contracts, enhancing security through cryptography, and potentially reducing costs for all stakeholders by eliminating intermediaries.
  • Securing a Competitive Edge: In today’s competitive landscape, the key to differentiation lies in ecosystem integration. Blockchain technology, as a shared infrastructure, allows businesses to build all-inclusive ecosystems where partners can collaborate freely. This fosters innovation and value creation across the network, creating a powerful “lock-in effect.” Early adopters who leverage blockchain’s potential to build robust ecosystem become pioneers, attracting partners and establishing a long-term competitive edge.

Financial Institutions and the “Reals” Revolution

Having explored the critical questions surrounding blockchain technology and the Reals, let’s dive into why financial institutions should participate in this evolving financial landscape. If they choose to embrace this revolution, how can these incumbents actively be involved? Furthermore, what are the key gaps they need to address to fully capitalize on the ‘Reals’ revolution?

1. The Rationales Why the Reals is a Strategic Imperative for FIs

The financial landscape is rapidly evolving, driven by innovative technologies like blockchain. To stay ahead of the curve and solidify their leadership positions, FIs need to embrace the “Reals” framework: Real Money, Real Assets, and Real Identity for the following reasons:

• • Building a Competitive Future with Existing Strengths:
    • Innovative Products and Enhanced Security: Blockchain technology allows FIs to leverage their existing expertise in security and data management. By focusing on Reals, FIs can develop innovative financial products with a foundation of trust and security. For example, HSBC’s pilot project in the UAE used blockchain technology in the know-your-customer (KYC) process. This initiative allows the secure sharing of verified KYC data between banks and licensing authorities in the UAE, which could simplify the onboarding process for the clients while maintaining data integrity. This focus on security fosters client adoption and positions FIs as trusted leaders in the evolving financial landscape.
    • Cost Optimization and New Revenue Streams: Reals-based solutions can unlock significant cost savings and create entirely new revenue streams. Streamlining processes and reducing transaction and settlement costs with blockchain translates to a more competitive edge, as seen with Franklin Templeton’s tokenized money market fund. With this initiative, tokenization opens doors to new investment opportunities for both existing and new clients, secondary trading possibilities and collateral use, generating additional revenue sources.

• • Securing Future Share of Wallet:

• • • Foundation for Seamless Integration: Building infrastructure that aligns with Reals allows for more effortless integration with future Web3 and blockchain applications. This establishes FIs as key players in the evolving ecosystem.

• • • Network Effect: Early adoption of Reals helps FIs cultivate a robust network of participants within their ecosystem. As new applications and services emerge, this network effect positions them for continued growth and relevance.

2. How can FIs Dive into Real

There are growing real-world use cases that leverage blockchain technology. Instead of listing potential use cases that FIs could explore in Real Money, Real Assets, and Real Identity, this article will discuss the framework we deem essential for FIs to consider. This framework will guide them in exploring and prioritizing which Reals use cases they should develop and launch.

Prioritizing The Reals Use Cases

To navigate the exciting possibilities of the Reals, FIs should consider and achieve these three aspects:

• • Problem-Solving Impact:

• • • Focusing on Real Problems: Identify pain points in existing money, asset, and identity management from within the organization and the clients. This could be slow cross-border payments, opaque trade finance, or cumbersome KYC processes
    • Aligning with Business Goals and Client Needs: How can a solution address strategic goals like cost reduction, revenue growth, or improved customer experience? Solutions built with Reals should leverage familiar and intuitive UX/UI, ensuring a smooth transition and minimizing the learning curve for the users
    • Assessing the Impact: Assess potential impacts in terms of the implementation in both financial and technical aspects

• • Technical Viability:

• • • Leveraging Existing Solutions: Can existing platforms or solutions be adapted to address the identified problem?

• • • Evaluating Client’s Resources: Consider client’s capability and willingness to adopt and implement blockchain technology

• • Regulatory Compliance:

• • • Compliance Considerations: Evaluate potential regulatory hurdles related to privacy, customer data protection, cybersecurity etc.
    • Risk Management: Balance the innovation with customer protection and financial stability

Balancing the Framework: Why All Three Factors Are Important

The framework we presented emphasizes the importance of considering all three factors – Problem-Solving Impact, Technical Viability, and Regulatory Compliance – to ensure successful Reals implementation. Here’s why each factor plays a crucial role:

• • Problem-Solving Impact + Technical Viability ≠ Success: Even if a use case addresses a real problem and is technically feasible, it could be an ‘illegal innovation’ if it clashes with the regulations.
  • Problem-Solving Impact + Regulatory Compliance ≠ Reality: A focus solely on solving a problem and adhering to regulations could lead to unrealistic solutions, like us in the ‘dreamland’.
  • Technical Viability + Regulatory Compliance ≠ Innovation: Focusing solely on the technology and regulations could lead to ‘overengineering’ solutions that do not address the core problems effectively. Sometimes, existing solutions with minor adjustments can achieve the desired outcome.

By taking all three factors into account, FIs should be prioritizing Reals use cases that deliver tangible and impactful benefits to the organization and/or clients, technically achievable while complying with current regulations to avoid potential roadblocks.

3. Bridging the Gaps

Despite the promise, there are challenges for FIs to address for a successful implementation of use cases in the Reals:

• • Regulatory Uncertainty: A major hurdle is the evolving regulatory landscape surrounding blockchain. The concerns can evolve around market stability, ownership of on-chain assets, investor protection, data privacy etc. FIs need a guideline from the relevant regulator(s) to move forward with confidence.
  • Technical Hurdles: The main technical issues are integration with legacy system, scalability limitations and standardization issue. Integrating blockchain solutions with existing core banking systems can be complex and expensive. In terms of scalability, FIs need to explore protocols that can simulate and handle real-world financial application. Due to differences in development standard, it is crucial to have a secure interoperability solution in place and/or develop a common standard for data formats and communication protocols.

• • Building Trust and User Adoption: Public education and intuitive user experience are the key to accelerate a mainstream adoption of blockchain and the use cases in the Reals. FIs need to invest in educating consumers about the benefits and security features of blockchain-based solutions, while developing user-friendly solutions to interact with decentralized applications.
  • Talent Acquisition and Skills Gap: Specialized skills such as blockchain architecture, smart contracts, data structures are required to launch and operate blockchain-based products/services. FIs need to invest in talent acquisition programs to attract developers and other business professionals, while providing training programs on blockchain basics to existing employees to narrow the skill gap.

Closing Thoughts: A Future Empowered by the Reals

With a clear understanding of the importance of Real Money, Real Assets and Real Identities powered by blockchain, the technology offers exciting possibilities and transformative path for FIs. By prioritizing Problem-Solving impact, Technical Viability, and Regulatory Compliance, FIs can create impactful results for all stakeholders. Consumers gain control and security, businesses find efficiency and innovation, and the economy thrives on inclusion and transparency. Overcoming the mentioned challenges can unlock immense potential, paving the way for a more secure, efficient, and innovative financial future.

Author: Wanwares Boonkong (Pin)

Editor: Panuchanad Phunkitjakran (Pook), Woraphot Kingkawkantong (Ping)

Reference:

• https://m-tiesler.medium.com/real-world-use-cases-for-blockchain-technology-b97c573b3253
  https://www.forbes.com/sites/forbestechcouncil/2024/01/17/how-blockchain-revolutionizes-data-integrity-and-cybersecurity/
• https://therecord.media/cybercriminals-stole-over-1-billion-from-crypto-funds-2023
• https://blogs.worldbank.org/en/trade/greasing-wheels-commerce-trade-finance-and-credit#:~:text=Globally%2C%20the%20trade%20finance%20%E2%80%9Cgap,US%246.5%20trillion%20in%202021.
• https://www.ledgerinsights.com/hsbc-goes-live-on-the-uae-know-your-customer-kyc-blockchain/
• https://www.coindesk.com/business/2024/04/25/franklin-templeton-upgrades-380m-tokenized-treasury-fund-to-enable-peer-to-peer-transfers/
• https://www.mckinsey.com/industries/financial-services/our-insights/blockchain-and-digital-assets
• https://hbr.org/2022/01/how-walmart-canada-uses-blockchain-to-solve-supply-chain-challenges

Image generated by Gemini

Given the rapid changes in customer behavior, incumbent financial institutions must adapt to remain competitive in the market. Customer touchpoints and experiences become increasingly important factors in differentiating their position.

As digital banking has become increasingly common, customer expectations for timely support have understandably risen. However, financial institutions often struggle to achieve deep consumer engagement solely through mobile apps. For more complex products like investment funds, bonds, or intricate investment vehicles, human assistance remains crucial. Human advisors can guide customers, understand their risk appetite, and find suitable wealth products. Some customer requests cannot be done purely by clicking the button. It is important for financial institutions to understand the “situation” and “unique requirement” customers are demanding, while being attuned to the customer emotional state, and response accordingly. Since customers are active on various channels, effective communication and engagement method are vital in capturing their attention and meeting their expectations.

Therefore, banks must consider leveraging new technology and automation, on top of already existing tools, to boost customer satisfaction while streamlining operational costs. One promising solution is “conversational banking.”

Conversational Banking: An Evolution of AI for Financial Services

Conversational banking is a rapidly growing trend in the financial services industry, improving the way banks interact with customers by leveraging technologies such as artificial intelligence (AI), in conjunction with existing chatbot, to deliver more personalized and accessible banking experiences.

Customers nowadays often expect faster response time from their preferred communication channels. Chatbots are among the most common applications to expand banks’ service time to 24/7 and the earliest applications of conversational AI in banking. They are on the edge to help customers from mundane financial activities such as transferring money and checking account balances, to less straight-forward activities such as account opening, managing investment portfolio, and negotiating credit card payment terms, all without the need to tediously scan through websites or apps or wait on hold for a call center.

However, due to their constraints around the ability to comprehend only specific use cases and precise keywords, chatbots often force customers to constantly adapt their language to structured commands or predefined phrases, which can be frustrating for customers and loss of opportunity for banks. Based on this pain point, conversational banking offers a more natural ways for customers to interact with banks, and the engine adapts over time using machine learning to learn from past interactions and make the chatbots smarter.

In short, conversational banking takes traditional chatbots to the next level. It empowers chatbots to learn from past interactions and anticipate user needs, while able to understand natural language used by customers, resulting in a more engaging experience through “human-like” interactions. These “smarter chatbots” leverage various technological tools to enhance their capabilities and achieve more for both customers and banks.

The Tech: Automation of Human-like Interactions from Rich Data

Unlike static chatbots, conversational AI learns from every interaction with its users. Each conversation feeds its machine learning (ML) engine, enabling it to handle advanced terms, local slang, and dialects. Behind conversational AI, there are various supporting technological tools to make the bot learn, analyze and response.

In a non-technical term, the AI first tries to understand what a user is asking, then chooses the best way to respond, and finally makes sure it sounds natural. If the user is using voice, it also needs to understand what the user says, both message and tone, and replies clearly and empathetically.

As illustrated above, conversational banking leverages a range of technologies to resolve customer queries including Natural Language Understanding (NLU), Dialog Management, Natural Language Generation (NLG), and Automatic Speed Recognition (ASR) (details of each tech are noted in the table below). Each tool works together to make the bot function. NLU and NLG are parts of Natural Language Processing (NLP).

NLP offers a range of powerful capabilities, including speech recognition, speech-to-text conversion, and text-to-speech synthesis. Furthermore, NLP can now even identify emotions in text or speech, adding a new level of understanding to human-computer interaction.

Importantly, the machine learning tool helps the bot analyze and learn from past conversations, and applies them to the conversation at hand.

Table 1: Bundled AI tools help the bot work and learn.

Due to technology like NLP and ML, AI has become smarter and more human-like.  Additionally, another key development driving this progress is the data used to train models. In the past, traditional single-model AI relied on a single source or type of data for specific tasks such as scribing texts from the internet to teach AI. However, multimodal AI ingests and processes data from multiple inputs like text, video, images, and speech. By combining relevant data from various sources (not limited to text), AI has vast amounts of information to learn and analyze, which could be put together into smarter and more engaging responses.

The Promise: What Human-Like Chatbots May Bring to the Market

The way financial institutions interact with customers is constantly evolving. Conversational banking disrupts these traditional methods by leveraging new technology to respond to ever-changing customer behavior. For the purpose of this article, we can examine the impact of technological advancement on how banks interact with customers.

Tech Advancement: From Static to Dynamic Scripting

As mentioned earlier on chatbots vs conversational AI, the technology advances from bots responding to simple questions or requests, based on pre-analyzed models and logic, to adding some sorts of analytics and personalization. Today, due to reduced technology costs of AI development and cloud computing, real-time analytics is made possible. Bots could gather data from various sources in real-time to analyze and predict customers’ next requests.

The form of conversational AI is enabling banks to move away from making simple requests like transferring money, to provide deeper analysis such as spending graphs with monthly comparisons, to personalized solutions for each user, and real-time monitoring to report and support. Now the next move is Anticipative Interaction. AI anticipates customers with specific needs before they even reach out. Even before customers get in touch, an AI-supported system can anticipate their likely needs and generate prompts for the agent (as co-pilots) or for decision maker to fabricate pre-made solutions. For example, the system might flag that the customer’s credit-card bill is higher than usual, while also highlighting minimum-balance requirements and suggesting payment-plan options to offer. If the customer calls, the agent can not only address an immediate question, but also offer support that deepens the relationship and potentially avoids an additional call from the customer later.

Interactions: From Reactive to Anticipative

1. Where or Channels

Customers are migrated from traditional branches to digital channels. Traditionally, the main interaction between banks and customers is between bank branches and ATMs. Today, mobile banking is becoming a main point of contact, therefore, call centers and chatbots are playing larger roles in customer support.

Soon, chatbots and conversational AI will expand to every touchpoint of clients. The service could be integrated with different apps (rather than mobile banking apps) and channels (even branches to lower headcount costs) that users are interacting with. One example is the accessibility through third-party messaging services and social media platforms to improve the experience to customers. A customer may see a news recap on political conflict in a neighboring country on social media, forward it to the bank’s official account chat-box, and inquire about next steps on mitigating the effect on his/her investment portfolio. Customers would benefit from 24/7 support, reduced waiting time, and more personalized responses.

2. What or Use cases

In the past, banks primarily used conversational AI for customer verification (Know Your Customer or KYC) and handling simple transactions like balance checks and money transfers. However, as advanced conversational AI can learn from experience, it opens doors for new use cases in the financial service industry.

The use of data can supercharge the bots’ analysis, providing deeper customer insights and fostering broader financial service integration. For example, conversational banking could support pre-qualification for loans by analyzing non-financial data such as voice to predict credit scores. It could also remind customers about upcoming payments based on their spending patterns or approve loan extensions using past repayment and social media data.

Looking to the future, conversational banking is expected to extend beyond banking products and services. By connecting data with third-party platforms, it could help customers improve their quality of life. Financial institutions could partner with other service providers such as exchanging bank loyalty points for airport pickups, or automatically ordering groceries for home delivery, or identifying surges in utility bills and then suggesting a financing plan for solar rooftop purchase. These services could be learned from past transactions and predicted by AI.

3. How or Humanness

Conversational banking evolves chatbots beyond simple FAQ responses. Initially, they are trained with pre-defined prompts and answers. As they progress, they learn industry-specific vocabulary and become more flexible in understanding customer keywords to find relevant information.

Today, the level of analytics has greatly improved. AI will be built and currently have built to be more like humans with the learning of emotions and tone of voice when dealing with customers. Additionally, as the AI learns through multiple conversations, it increases the level of personalization and ability to support individual clients.

For example, in various use cases like cross-selling new products and collecting loan payments, AI will identify the right tone of voice such as softer and higher pitch voice to interact with each customer. Adversely, AI would analyze the emotions of customers during the interaction to identify suitable responses or when to escalate the conversation to the human support team. To drive a personalized experience, servicing channels are supported by AI-powered decision-making, including speech and sentiment analytics to enable automated intent recognition and resolution.

Why Now?

AI adoption poses benefits in terms of cost reduction, improved customer satisfaction, and increased competitive advantage. To stay competitive, embracing AI is no longer a choice, but a necessity. Leading institutions are already leveraging advanced AI to serve customers, empower employees, and secure their market share.

Higher automation reduces costs and improves customers’ satisfaction through operational efficiency, minimizing errors and optimizing resource utilization. Conversational banking further enhances customer experience by streamlining human resource allocation, reducing response times, and improving account access and security, while also reducing fraud. Conversational AI plays a key role by integrating data from various teams. This comprehensive data view empowers institutions to manage resources more effectively and ensure compliance with the evolving market and regulatory landscape.

HSBC’s use of conversational banking serves as a great example. Launched in June 2018, HSBC’s AiDA chatbot is used to respond to clients’ requests via instant messaging, reducing the cost related to calls by 90%. In February 2021, HSBC used a chatbot powered by AI to provide instant pricing and analytics for foreign exchange options, making complex trading more accessible and efficient.

Leading financial institutions are increasing the use of advanced AI technologies. McKinsey’s Global AI Survey reveals that nearly 60% of financial-service sector respondents have already embedded at least one AI capability into their operations. This digital transformation is taking shape through:

• Conversational bots for basic servicing requests
• Humanoid robots in branches to serve customers
• Machine vision and natural-language processing to scan and process documents
• Machine learning to detect fraud patterns and cybersecurity attacks from conversations

Technology helps speed up the tedious process and push financial institutions ahead of other banks and Fintechs. The new tech adoption will become standard practice or baseline in the eyes of customers.

Consumers are increasingly shifting towards digital channels, favoring mobile banking for simple services, and reducing visits to physical branch. This behavior shifts emphasizes the need for banks to adapt. Mobile banking, pioneered by commercial banks in Thailand, has been a successful driver of the digital trend. Presently, over 80% of bank clients have and regularly use mobile banking applications. In addition, government initiatives such as PromptPay and the G-Wallet policy, aimed at boosting domestic spending, have further accelerated the adoption of digital payments and mobile banking. A 2022 Mastercard survey revealed that a staggering 94% of Thai consumers now use digital payments. With Gen AI gaining popularity in public eyes, commercial banking is expected to play a an increasingly significant role in supporting banks’ customers in the near future.

Non-banking businesses are entering the banking space. Banking business is now embedded in a wide range of software and applications (See more about Embedded Finance here). One significant threats to banks is the emergence of  “Super Apps” (See more about Super Apps here). These Super Apps integrate various financial services, including payments, and in some cases, lending, and insurance, potentially becoming one of the main operating businesses and posing a threat to incumbent banks. They disrupt traditional methods of offering new banking products and services and may soon seek to expand their presence and involvement in financial services on a larger scale. As a result, financial institutions will need to reassess how they participate in digital ecosystems and leverage AI to unlock the untapped data potential for competitive advantage.

Things To Be Cracked

Image generated by Gemini

Conversational AI could help boost banks’ presence and support clients; however, the technology presents challenges and hidden costs. New tech adoption is poised to disrupt how banks manage and operate. Incumbent players face a balancing act: ensuring agility and flexibility for competition while maintaining security and compliance to secure trust as financial service providers.

1. Infrastructure Readiness:

Implementing conversational banking requires robust computing power and flexibility to support real-time analysis. However, legacy core banking systems are often difficult to modify. Additionally, fragmented data across different teams hinders the analysis of relevant data and timely generation of recommendations. Most importantly, transitioning to a new infrastructure and ongoing computational requirements can incur significant costs.

2. AI and Talent Management:

A successful AI strategy requires a clear roadmap for both technology and talent. Currently, a major challenge is the lack of standardization in conversational AI adoption. Organizations are still exploring the best ways to integrate technology. The strategy should encompass both the technical process of tech implementation (building, testing, deploying, and monitoring) and talent development to uplift existing employees to develop and maintain new products and services. Ultimately, the strategy must identify and develop use cases where conversational AI can transform customer journeys, leading to defined outcomes such as real-time client support, tailored service with insightful data, improved customer lifetime value, and lowered operating costs.

3. Regulatory and Ethical Considerations:

Disruptive technologies often raise regulatory concerns, particularly for financial institutions where reputation and stability are critical. This is especially true for conversational AI, which handles sensitive personal and financial information. Personal data management and changing regulations should be closely monitored when adopting conversational AI. Additionally, there are ongoing developments in the framework to help reduce human biases (such as social profiling) imposed on AI models. For example, Thailand’s Electronic Transaction Development Agency (ETDA) has published AI Governance Guidelines with hopes of bolstering domestic ethical AI development and adoption.

Closing Thoughts

Conversational AI offers numerous benefits in enhancing the customer journey. However, as with any technology, it comes with associated costs. Financial institutions should be aware of the limitations of AI technology and carefully balance the costs and benefits of adopting it. Ultimately, the success lies in the hands of banks that could identify where best to implement new technology and where its implementation might not be the most suitable solution.

Author: Panuchanad Phunkitjakran (Pook)

Editors: Supamas Bunmee (Jae), Woraphot Kingkawkantong (Ping)

Image by UNESCO Bangkok

The playing field of education shouldn’t be tilted by wealth, but in a world where socioeconomic status (an individual’s social standing based on economic status) casts a long shadow, it often is. While differences in race, gender, or nationality can shape life trajectories, disparities in income paint an even starker picture. In Asia-Pacific, according to Asia-Pacific Social Science, for instance, the richest 25% of households enjoy opportunities 13 times greater than the poorest 25%. Enter the DEI (Diversity, Equity, and Inclusion) movement, a beacon of hope aiming to bridge such divides. But what does DEI look like in a country like Thailand?

Here, the education gap reigns supreme. FleishmanHillard Research (2023) found it the top DEI priority. Thailand’s educational landscape is booming. International schools sprout like mushrooms, even going public, while top schools boast cutting-edge tech classes like blockchain and AI. Yet, only those with deep pockets can access this gilded future, evidently shown by Thai students’ very low on PISA index in every factor. This ironic reality – where advancement widens the gap instead of closing it – demands immediate attention.

This article delves into the heart of this matter, dissecting how socioeconomic status breeds educational disparities, followed by our thesis of how we can collectively address these disparities. Then, we will also make distinction between two important concepts, Education Inequality and Education Inequity, and argue that solving Education Inequity is most paramount. We’ll explore the role of EdTech, a potential equalizer, and alongside other diverse stakeholders can collaborate to bridge the educational divide. Join us as we embark on this critical journey, where the future of Thailand’s children hangs in the balance.

What Is DEI and What Is Its Relevance To Education?

Diversity, Equity, and Inclusion (DEI) is a part of the ESG movement, specifically the Social part, that aims to create a world where everyone is equally worthy, able to strive, and lives in harmony despite all differences. Though the concept of DEI originated from the issue of race and gender, it has been developing to cover all other aspects including education, political beliefs, and socioeconomic status. Let’s get to know each component:

• Diversity: Acknowledging the richness of human variation, encompassing not just visible traits like race and ethnicity but also invisible factors like socioeconomic background and educational attainment.
• Equity: Leveling the playing field by providing targeted support and resources to bridge the gap between different groups. This goes beyond equal access to ensuring equal outcomes.
• Inclusion: Creating a sense of belonging and value for everyone, regardless of their background. This fosters a sense of community and empowers individuals to contribute their unique perspectives.

By understanding these interconnected elements, we can see how DEI directly addresses the challenges of education equity, urging us to recognize the individuals’ different background and circumstances (e.g., socioeconomic status) and provide equitable resources to ensure the same educational outcome. It’s about dismantling barriers and fostering a system where every student, regardless of their socioeconomic status, has the opportunity to reach their full potential.

Beyond Equality: Why Education Equity Among Socioeconomic Status is Thailand’s DEI Imperative

While the DEI movement in the West often focuses on race and gender, in Thailand, it takes a different form. As FleishmanHillard research (2023) reveals, a staggering 32% of Thai people identify education inequality as the most pressing DEI concern, placing it at the pinnacle of the DEI issues that need to be addressed. This is no mere coincidence.

Source: FleishmanHillard Research, 2023

While “education equality” aims to provide equal resources to all students, it doesn’t guarantee equal outcomes. This is where “education equity” steps in. It strives to ensure that despite differing backgrounds, all students reach similar educational benchmarks and are equipped to compete in the job market and have an equal chance for social mobility.

Think of it this way: providing every student a book (equality) is meaningless if some lack the support or environment to read effectively (equity). Education equity addresses these disparities by offering targeted resources and support, such as scholarships and financial aid workshops, specifically for students from low-income families.

Source: McGraw Hill PreK-12

In fact, when we take a look at what factors prevent Thailand from achieving education equity, research by Asia-Pacific Social Science Review (2022) reveals that while various factors like language, disability, and location contribute to education inequity, socioeconomic status consistently ranks as the most impactful component in Thailand. The parents’ socioeconomic status has played a significant role in children’s opportunities in higher education. This critical issue deserves attention for two key reasons:

• Sizable Affected population: According to KKP research (2021), the richest 10% own over 77% of the country’s wealth. Given such a high level of wealth disparity, a significant portion of the population is struggling to afford quality education for their children.
• The Persistent Loop of Poverty: Limited education often leads to lower income, perpetuating the cycle of poverty. As the Organisation for Economic Co-operation and Development reports, a university degree can result in wages nearly 2.5 times higher than a lower secondary degree. Without education equity, this gap widens with each generation, trapping individuals in a cycle of disadvantage.

In summary, achieving education equity among socioeconomic status is not just a moral imperative; it’s an economic necessity for Thailand’s future.

Unequal Playing Field: Navigating Education and Employment by Socioeconomic Status

Socioeconomic status casts a long shadow on Thai education and employment opportunities, creating distinct tiers with varying access to resources and success. While acknowledging the complexity of such categorizations, we can broadly divide Thai society into three segments based on their educational and economic realities: the Privileged, the Mainstream, and the Strugglers.

The Privileged: This segment enjoys abundant resources and opportunities. Their families can afford quality education, extracurricular activities, and skill development, often equipping them with advanced qualifications and specialized knowledge. This translates to access to high-paying jobs in professional fields and the potential to further accumulate wealth.

The Mainstream: This segment comprises a significant portion of the population with sufficient resources to attain basic education and essential skills. They are actively engaged in the job market, securing skilled positions and earning enough to cover their needs. While financial security is attainable through hard work and dedication, upward mobility within this group can be challenging.

The Strugglers: This segment faces significant economic hardship and limited resources. Meeting basic needs consumes their energy and income, leaving little room for education or skill development. They often rely on low-paying jobs with minimal opportunities for advancement, perpetuating a cycle of poverty. This lack of access to quality education and resources severely hinders their ability to break free from this cycle.

The Urgency of Equity: Empowering the Strugglers

While all groups navigate challenges, the Strugglers face a unique predicament. Without external support, their ability to break the cycle of poverty through education is severely restricted. To put this simply, they lack the means to access the tools needed for upward mobility on their own.

By focusing on bridging the educational gap for the Strugglers, Thailand unlocks the potential of a large segment of its population. This, in turn, fosters a more equitable society with a broader tax base, increased productivity, and a more just distribution of wealth. Ultimately, investing in the Strugglers is not just an ethical imperative, it’s a strategic investment in the future of Thailand.

The Path to Equity: A Three-Pronged Approach for Thailand’s Strugglers

Bridging the educational gap for Thailand’s Strugglers requires a multi-faceted approach that tackles the Strugglers’ unique challenges. Here, we propose a three-pronged strategy involving various stakeholders to pave the way for educational equity:

1. Freeing the Strugglers from Financial Burden: Kick-starting Successful Learning Journey

The financial hardship casts a long shadow on a Struggler’s educational journey. Parents grapple with the impossible choice between immediate survival and investing in their children’s future. This burden manifests in several ways such as child labor and parental pressure for children to contribute financially, and limited ability to afford financial resources. To address this issue, there are several potential areas to be addressed such as:

• Targeted financial assistance: Scholarships, grants, and loan forgiveness programs, either channeled directly to the families or schools, specifically designed for Strugglers can alleviate the immediate financial pressure of school fees, uniforms, and educational materials.
• Conditional cash transfers: Providing financial assistance directly to families, on the condition that their children attend school regularly, can incentivize education and reduce child labor.
• Subsidized childcare and after-school programs: Freeing up parents’ time by providing affordable childcare and after-school programs can allow them to work without sacrificing their children’s education.

2. Uplifting the Landscape: Building Equitable Learning Environments

The next step is to address the disparities in educational resources and infrastructure. This requires a concerted effort to ensure Struggler schools are equipped to provide quality education on par with the Mainstream. This disparity manifests in several ways such as teacher quality, limited and out-of-date equipment and facilities, lack of community support, and obsolete curriculum and teaching materials. To bridge these disparities, below are some areas that would benefit from immediate intervention:

• Targeted investment in rural and underserved schools: Increased funding and resource allocation specifically for schools catering to Strugglers can ensure they have access to qualified teachers, modern technology, and up-to-date resources.
• Teacher training and support: Providing ongoing training and professional development opportunities for teachers in underserved communities can equip them with the skills and knowledge necessary to effectively support Strugglers’ learning.
• Curriculum reform: Integrating real-world skills and relevant job market trends into the curriculum such as coding, basic technology knowledge like Blockchain and AI, or sales and presentation skills, can prepare Strugglers for future success and make learning more meaningful.

3. Empowerment and Personalization: Tailoring Education to Individual Needs

At the heart of any successful learning journey lies a strong internal drive to learn and succeed. For Strugglers, it is often hard to imagine life beyond the status quo given their limited exposure to role models and information about diverse career paths. Additionally, witnessing their parents’ struggles can lead to self-doubt. Negative experiences or societal stereotypes can also lead to feelings of inadequacy, hindering Strugglers’ belief in their ability to achieve their goals. Below are some solutions that can address the issue:

• Mentorship and career guidance: Connecting Strugglers with mentors from similar backgrounds or experienced professionals can provide invaluable advice, role models, and networking opportunities, helping them navigate career choices and access job markets.
• Internship Opportunity: Providing Strugglers with a field to exercise their classroom knowledge in real-life situations not only strengthens their skills but also increases their recruiting opportunities.

Educational equity demands a move beyond one-size-fits-all approaches. Each Struggler student has unique goals, learning styles, and aspirations. This diverse landscape requires personalized learning pathways based on their learning style and goals, personalized mentorship and career guidance, and targeted skill development programs suited for excelling in the job market.

• Adaptive learning platforms: These platforms personalize learning pathways based on individual student progress, strengths, and weaknesses, ensuring efficient knowledge acquisition and catering to diverse learning styles.
• Micro-credentialing and skills-based learning: Offering bite-sized, skill-focused courses allows Strugglers to acquire relevant skills in short periods, even if they cannot pursue full-time degrees. This can be particularly helpful for those seeking immediate employment opportunities.

Building Bridges, Not Walls: A Collaborative Approach to Education Equity in Thailand

Bridging the educational gap for Thailand’s “Strugglers” demands a collective effort, not a solitary sprint. Each stakeholder in the education ecosystem plays a crucial and unique role in dismantling barriers and building a future where every child, regardless of background, has the chance to thrive. The discussion below provides a general frame of thought on how each stakeholder could mainly contribute. Much of what is being described below has already been done sparsely and uncoordinatedly, but Thailand as a nation can do so much better to ensure equitable education for the Strugglers.

Governments act as architects of supportive infrastructure. Firstly, infrastructure can be leveled by equitable resource allocation, either in the form of fiscal budget allocation or tax incentives for other stakeholders to contribute, ensuring that Strugglers in rural and underserved schools have access to qualified teachers, modern technology, and up-to-date resources. For more thoughts on closing the digital inequality, please visit Beacon VC’s article here. Secondly, infrastructure can be future-proof by integrating real-world skills like coding and AI into the curriculum preparing Strugglers for the job market and making learning more relevant to their aspirations. Lastly, infrastructure can be more inclusive by implementing programs that provide financial assistance to families in exchange for their children’s school attendance can incentivize education and reduce child labor. This requires close collaboration with social welfare ministries and community organizations for effective implementation.

Financial institutions act as fuel for change. Leveraging the financial capability, access they have to Thai communities, and the amount of human resources they have, financial institutions can catalyze the transition at both macro and micro levels.

At the macro level, financial institutions can join hands with several stakeholders, such as government and NGOs, to structurally build equitable education systems, through targeted scholarships and loans designed specifically for Strugglers, families can prioritize education without sacrificing immediate needs. Additionally, financial institutions can also channel investments into areas that would advance solutions tailored to Strugglers’ unique challenges, such as EdTech’s affordable learning platforms, adaptive online learning technologies, or micro-loan programs for schools. Financial institutions can also play an active role in shaping financial literacy for Strugglers about budgeting, saving, and responsible credit management can empower them to make informed financial decisions regarding their children’s education.

At the micro level, financial institutions can have a direct and profound impact on individual Strugglers who have the potential to excel. Through specially designed initiatives for Strugglers like internship/ apprenticeship programs or mentorship and career counseling programs, in partnership with local schools or vocational institutions, Strugglers can get inspiration and obtain relevant skills within the field and inspiration to push their career forward. Inversely, financial institutions will have direct access to a talent pool that is trained specifically for their unique organizations’ business and operational requirements.

NGOs and surrounding communities act as networks of support. At the national or municipal level, using their collective voice, NGOs and communities can advocate for the awareness of Struggler’s situation and raise public support for policy changes. At a community level, providing affordable daytime childcare and after-school programs can free up parents’ time and allow them to work without sacrificing their children’s education. Lastly, at the individual level, there’s also an opportunity for mentorship and career guidance programs to connect Strugglers with mentors from similar backgrounds or experienced professionals, providing invaluable guidance and role models.

EdTech startups act as architects of personalized and accessible learning. At the heart of education equity, there’s an important recognition that all students learn differently, at a different pace, and for different purposes.

On one hand, EdTech startups are well equipped to address this through the ability to tailor learning experiences down to different individuals using AI/ML in their adaptive learning platforms, tailoring courses based on individual strengths and weaknesses. Micro-credentialing and skills-based learning allow Strugglers to pick-and-choose relevant skills to acquire in short periods, even if they cannot pursue full-time degrees.  On the other hand, EdTech startups can also assist schools to partially overcome resource constraints in teaching or tailoring students’ education pathways, starting from solutions as fundamental as helping teachers track their students’ homework to tools to run remote classrooms for students in hyper-rural areas.

By working together, each stakeholder becomes a vital link in the bridge, not a barrier on the path. Only through collaborative action can we dismantle the walls of inequality and build an education system that truly empowers Strugglers to reach their full potential. In the next section, we’ll zoom in on the Thai EdTech landscape, examining specific examples of how these innovative tools can tailor learning, dismantle barriers, and empower Strugglers on their path to success.

Bridging the Gap: How EdTech in Thailand Can Contribute Through Personalized and Accessible Learning

Source: @terrynut, Medium

Edtech in Thailand has been expanding in line with global trends, reflected by the rise in number of users especially after the Covid-19 period. Digital learning platforms and e-learning solutions were becoming increasingly popular, offering a range of subjects and flexible learning options. This aligns with the findings of a survey conducted by Kasikorn Research Center in April 2021, which found that 96% of respondents anticipated a higher inclination towards using EdTech and online learning. This is especially true for regular employees aiming to enhance their skills and make productive use of their free time.

Riding the boom, EdTech startups have the potential to play a crucial role in achieving education equity, particularly for students facing socioeconomic disadvantages, through 1) Uplifting the Landscape: Building Equitable Learning Environment, and 2) Empowerment and Personalization: Tailoring Education to Individual Needs. Let’s revisit the framework for education equity and explore how EdTechs are already tackling the issue:

Uplifting the Landscape: Building Equitable Learning Environment

• Democratize learning Materials: Ookbee‘s digital content platform makes reading materials more accessible and affordable for students from various backgrounds.
• Enhanced Learning Management Systems: SchoolBright empowers educators with tools for managing virtual, hybrid, and in-person classrooms, improving accessibility for students in rural areas.
• Teacher upskilling: Inskru is an online platform that aims to connect, inspire, upskill, and empower teachers across Thailand on various topics like coursework management, in-class activities, and student engagement. Starfish Labz curates short courses that aim to equip teachers with tip and tricks to be more effective in the classroom.

Empowerment and Personalization: Tailoring Education to Individual Needs

• Tailored Online Career Counseling: Platforms like WE Space and Dynamic School Thailand guide students towards informed career choices by offering assessments and suggesting opportunities aligned with their interests and strengths. They also provide access to relevant courses and workshops, fostering a real-world understanding of career paths.
• Bite-size Online Learning: Platforms like OpenDurian offer affordable online tutoring by connecting students with qualified tutors, regardless of location. Skillane and FutureSkill cater to diverse needs by providing access to up-to-date subjects not always available in schools.
• Learning Analytics for Personalization: BrightBytes leverages data on student performance and engagement to provide personalized learning experiences, identify individual needs, and track progress, offering valuable insights to educators. Starfish Class helps teachers identify unique talents and potentials of the students to be able to support accordingly.
• Internship Opportunity Platforms: เด็กฝึกงาน and JobsBD connect students with internship and job opportunities across industries, allowing them to gain practical experience and explore career options.

While EdTech has the potential to revolutionize education and promote equity, its journey in Thailand encounters 2 major challenges that hinder its widespread adoption:

1. Familiarity with traditional teaching methods. Resistance from schools is rooted from the concerns about difficulty in integrating technology into existing curriculum and training teachers in new methods. Thus, they decide to stick with familiar traditional approaches.
2. Budget constraints. Schools, especially public ones, struggle with the initial and ongoing costs of acquiring and maintaining hardware, software, and internet infrastructure. This burden extends to individual families, who may not be able to afford subscriptions or devices that would grant access to EdTech solutions.

These challenges highlight the need for a collective effort to bridge the education gap. Governments must invest in infrastructure and training, schools need to embrace innovation, and EdTech startups must offer affordable solutions. This collaborative approach is crucial for EdTech to effectively transform classrooms and empower students from diverse backgrounds, paving the way for educational equity in Thailand.

Closing Thought: A Future Built on Equity, Not Equalities

Education inequity casts a long shadow in Thailand, yet a collective yearning for change pulsates beneath. The chasm between the Privileged, Mainstream, and Strugglers reveals a stark truth: education is not a mere ladder, but a complex ecosystem demanding equal outcomes, not just inputs.

EdTech emerges as a beacon of hope in this landscape. Its potential to personalize learning, bridge access gaps, and dismantle socioeconomic barriers can rewrite the narrative of Thai education. From online platforms to immersive experiences, these tools empower the Strugglers, the very students whose potential remains locked away.

But challenges stand as sentinels guarding this path. Traditional mindsets and tight budgets threaten to stall progress. To forge a new road, collaboration is key. EdTech startups must champion ease of use, affordability, and platform benefits. Financial institutions can bridge the gap with support, knowledge, and affordable financing. The government’s role lies in building robust infrastructure, promoting equitable resource distribution, and incentivizing innovation.

Through the Beacon Impact Fund, Beacon VC aims to propel Thailand towards educational equity, recognizing it as a crucial social pillar within the ESG framework. The fund aims to provide support and network to fast-growing startup companies that aim to excel education equity and democratize access to opportunities across the country.

This journey towards education equity requires not just technology, but a collective will. When EdTech’s tools align with innovation, collaboration, and a focus on the most vulnerable, the Thai educational landscape can blossom into a tapestry of diversity, equity, and inclusion. It is a landscape where every learner, regardless of background, can unlock their full potential and paint their bright future.

Authors: Woraphot Kingkawkantong (Ping) , Pobtawan Tachachatwanich (Pob)

Editors: Supamas Bunmee (Jae) , Wanwares Boonkong (Pin)