New Challenges in Financial Regulation: Green, Digital and Artificial Intelligence
On May 19, a forum titled “New Challenges in Financial Regulation: Green Finance, Digital Finance, and Artificial Intelligence” was held. Several experts gathered to discuss strategies for addressing new challenges and opportunities arising from the development of green finance, digital finance, and artificial intelligence, and engaged in in-depth discussions on balancing financial innovation with risk prevention.
The forum was moderated by Li Yao, Member of Standing Committee of the 14th CPPCC National Committee and Member of Economic Affairs Committee of CPPCC National Committee. Andrew Sheng , former Chairman of the Securities and Futures Commission(SFC) of Hong Kong; Wang Xian, Vice Chair of the Tsinghua University National Institute of Financial Research and Director of Research Center for Listed Company of Tsinghua PBCSF; Lv Zhongtao, former Chief Technology Officer of the Industrial and Commercial Bank of China; and Enoch Fung, Chief Executive Officer of the Hong Kong Academy of Finance (AoF) and Executive Director of the Hong Kong Institute for Monetary and Financial Research (HKIMR), shared their perspectives and engaged in a discussion.
Li Yao, Member of Standing Committee of the 14th CPPCC National Committee and Member of Economic Affairs Committee of CPPCC National Committee
Moderator Li Yao noted that the green transition, digital transformation, and the application of artificial intelligence are intertwined, reshaping the drivers of growth and the landscape of the financial sector, while also bringing about profound changes in risk structures, regulatory logic, and regulatory approaches. Striking a balance between innovative development and risk prevention and control, and building a modern financial governance system that is safe, efficient, inclusive, and green, are critical challenges we must address. Li Yao invited the guests to share their perspectives from multiple angles and engage in an in-depth discussion on the new challenges and opportunities brought about by technological innovation and the application of AI in the financial sector, as well as their impact on commercial banks and traditional regulation; pathways for the high-quality development of green finance; the balance between innovation and risk regulation; and Hong Kong’s risk regulation practices.
Andrew Sheng, former Chairman of the Hong Kong Securities and Futures Commission
China’s promotion of inclusive green development aligns with the direction of the United Nations Sustainable Development Goals. Through green and inclusive development and large-scale infrastructure investment, China has gained valuable experience in promoting income growth, expanding employment, and improving development conditions. It has accumulated extensive practical experience in areas such as new energy development and green project construction, providing reference cases for the global green transition. Currently, the implementation of global sustainable development goals still faces numerous challenges, including a lack of political consensus as well as shortcomings in project design, implementation capacity, and governance mechanisms.
Technology is disruptively transforming the tools and structures of the global financial system, requiring a shift from the traditional top-down, linear management approach to a mindset that places greater emphasis on resilience, collaboration, and ecosystem governance. Financial institutions must not focus solely on their own interests and narrow efficiency; they must also assume corresponding responsibilities in maintaining the overall stability of the financial system and safeguarding the public interest. Technology has transformed the speed, scale, and scope of financial operations. While it has enhanced systemic efficiency, it may also amplify the spread of crises and the spillover of risks. The key to managing this “double-edged sword” of technology lies in establishing appropriate governance and regulatory frameworks.
Artificial intelligence, machine learning, and deep learning are transforming the way regulators assess, monitor, and manage institutional and systemic risks. Through regulatory technology, regulators can gain a more comprehensive view of market information, identify risk correlations, and maintain the resilience and efficiency of the financial system. Regulatory thinking must shift from sector-specific, fragmented, and static management to comprehensive, networked, and dynamic governance. In an era of rapid digital asset development, it is necessary to establish a governance model with clear objectives, strong enforcement, and multi-stakeholder collaboration among the market, government, and society. Through prudent pilot programs, impact assessments, and large-scale implementation, we can gradually build a path for financial reform that adapts to complex ecosystems.
Wang Xian, Vice Chair of the Tsinghua University National Institute of Financial Research and Director of Research Center for Listed Company of Tsinghua PBCSF
The green industry is characterized by strong positive externalities, long-term investment horizons, and high risk, which leads to significant market failures in this sector. Early intervention in this field requires external forces to correct these market failures, such as the participation of multilateral international organizations. The establishment of the “Equator Principles” in 2003 effectively channeled government influence through market mechanisms, enabling micro-level actors to genuinely incorporate “green” considerations as constraints in their decision-making. How to transform government influence and the mobilization of international organizations into spontaneous green behavior within the market, and how to leverage financial instruments to mobilize market forces for green development, are critical challenges that the global community must collectively address. Focusing on the question of how to manage the relationship between government and market, the government plays a vital role in establishing standards and incentive mechanisms, while the market must integrate these government standards and requirements into the behavior and decision-making of micro-entities. Only then can a sustainable path for green finance be forged. In China’s green development, the government takes the lead while the market follows. By leveraging institutional advantages to unify standards and relying on the state-owned economy to establish a mechanism that integrates national strategy with market forces, the country has transitioned from rule compliance to standard-setting leadership. China has gradually emerged as a standard-setter and policy leader in this field.
Lv Zhongtao, former Chief Technology Officer of the Industrial and Commercial Bank of China 
Since 2026, there has been a flurry of releases of cutting-edge models, and the industry has experienced explosive growth. The government has elevated “AI+” to a national strategy, setting a target of 70% penetration for intelligent agents by 2027. AI is driving profound transformations in business models and talent structures across industries such as finance. Banks are transitioning from open banking to agent-based banking, reshaping their internal organizations while expanding inclusive services externally. The key lies in consolidating enterprise-specific knowledge into reusable assets and establishing a human-machine complementary model where “humans oversee decision-making and agents execute with precision.” By leveraging role-based agents, companies can advance “post + skill” development to form a closed-loop business ecosystem. Regarding security, the inherent hallucination risks of large language models have given rise to new threats such as prompt injection, cognitive misoperation, and tool poisoning. Enterprises must establish a comprehensive security defense system that is monitorable, traceable, intervenable, and auditable by focusing on six key areas: AI agent positioning, knowledge engineering constraints, multi-agent checks and balances, permission governance, offensive and defensive capabilities, and Full-process monitoring. At the regulatory level, it is recommended to implement categorized supervision based on the principles of “distinguishing between internal and external applications, gradual opening, and controllable responsibilities and rights.” For external services, a prudent and strict approach should be maintained, emphasizing assistance and guidance while discouraging decision-making on behalf of clients. For internal empowerment, innovative pilot programs may be conducted under safe and controllable conditions, with tool-based agents deployed in sandboxed environments to ensure operational traceability and clear accountability. Looking ahead, AI has evolved from a tool into a process hub and decision-making intermediary, with a multi-agent collaborative model taking shape. This requires us to re-examine the relationship between humans and machines as well as organizational structures, and to build a new, efficient, and controllable collaborative system. The application of AI in the financial sector is currently at a critical juncture where opportunities and challenges coexist. Only by grasping technological trends, fortifying security defenses, and refining the regulatory framework can we truly empower high-quality development.
Enoch Fung, Chief Executive Officer of the Hong Kong Academy of Finance (AoF) and Executive Director of the Hong Kong Institute for Monetary and Financial Research (HKIMR)
The 15th Five-Year Plan clearly states that Hong Kong should leverage its role as an international financial center and utilize the advantages of “One Country, Two Systems” to support the nation’s financial reforms and high standard opening-up. Amid a century of profound changes and a complex geopolitical landscape, financial institutions face multiple risk shocks, which have significantly impacted their business models and operations. Regulatory challenges are primarily reflected in four areas: First, maintaining financial stability, which is the foundation for technological innovation and development; second, the time lag between regulatory measures and evolving risks—as an international financial center, Hong Kong must balance regulatory consistency with the varying pace of development when formulating policies; third, risk management and corporate governance within financial institutions, particularly regarding the application of fintech, the responsible use of AI tools, and the assessment of climate change’s impact on asset quality; and fourth, data quality, ensuring the reliability of third-party data when relying on it. In terms of specific measures, the first is policy incentives: Hong Kong maintains foreign exchange reserves at more than 1.6 times the monetary base to fortify the defense of financial stability. Second is the regulatory sandbox, which provides a risk-controlled testing environment for financial technology innovation, helping the industry deepen its technical understanding while also enhancing the capabilities of regulatory authorities themselves. Finally, there is active participation in the relevant works of international organizations to ensure Hong Kong’s voice is heard in policy and standard-setting.
Efforts to ensure that financial services support the real economy, build a strong financial nation, and foster a robust international financial center, strong financial institutions, a strong financial talent pool, and strong financial regulation must be advanced at various levels. As the nation’s international financial center, Hong Kong’s future lies in leveraging its sound and reliable financial platform to effectively implement strategies of “bringing in” and “going global.”
Roundtable Discussion Session
Regarding the impact of technology on commercial banks and traditional regulation, Andrew Sheng pointed out that the viability of financial institutions and regulatory risks cannot currently be viewed in a linear, siloed manner. As banks move toward becoming universal institutions, mastering data and information is key to internal risk management, external regulatory oversight, and the specific analysis of opportunities and challenges. Regarding how China’s financial ecosystem regulation should be organized and implemented within the country’s financial regulatory framework, Andrew Sheng stated that banks should utilize intelligent models to analyze system operations and conduct model stress tests. Establishing a comprehensive model by China’s financial regulators could provide effective risk identification and regulatory tools for the global financial system.
Regarding China’s green development, our country has forged a development path with its own distinctive characteristics, leveraging its institutional strengths and vast application scenarios. Regarding how green finance can further advance toward high-quality development, Wang Xian pointed out that the core of the Equator Principles lies in transforming external constraints established by governments or multilateral organizations into internal drivers for financial institutions and real-economy enterprises, thereby better leveraging the role of market mechanisms. For green finance to achieve high-quality development, it must ultimately be driven by market forces to ensure the healthy and sustainable development of both green finance and green industries. In the future, it will be necessary to enhance the motivation of market entities to participate in green finance through measures such as refining regulations, improving information disclosure, and establishing carbon pricing mechanisms. This will enable financial institutions to influence their own decision-making through market pricing, which in turn will be transmitted to the green investment and low-carbon transition decisions of real-economy enterprises.
In discussions regarding the definition of the boundaries of “core decision-making authority,” Lv Zhongtao pointed out that this is a matter requiring careful balancing, with a focus on issues such as classification and categorization, the determination and attribution of responsibility, and the maintenance of human expertise. Regarding the discussion on how to provide sufficient regulatory sandbox space for “moderate trial and error by AI agents” while strictly adhering to risk management bottom lines, he stated that regulatory sandboxes are a crucial mechanism in balancing regulation and innovation. Control interfaces for AI systems must be retained. It is an objective reality that regulatory rules lag behind innovation to some extent; therefore, enterprises must fulfill their primary responsibilities for compliance and risk management. Second, regulatory sandboxes must improve efficiency to avoid unclear boundaries and prolonged stays. Third, when designing sandbox mechanisms, regulatory coordination must be strengthened to prevent gray areas and regulatory loopholes caused by inconsistent regulatory standards.
Regarding Hong Kong’s approach to regulating and promoting fintech, how Hong Kong’s green finance and fintech measures can better serve the nation’s green development and sustainable development goals, and the role of technology in driving the overall development of green finance, Enoch Fung outlined Hong Kong’s regulatory philosophy and practices, emphasizing the prudent promotion of innovation and the deepening of fintech development while upholding financial stability as a bottom line. He introduced the Hong Kong Monetary Authority’s “Fintech 2030” blueprint, which encompasses four key areas summarized as “DART”: Data, Artificial Intelligence, Resilience, and Tokenization. The blueprint adheres to the fundamental regulatory principle of “same activities, same risks, same regulation,” and further refines this framework by integrating the “DART” approach. Regarding green finance, Hong Kong faces both the opportunity presented by the regional shift in global green transition finance and the risks posed by short-term climate change. Its response primarily focuses on two regulatory measures: first, establishing a common classification framework; and second, actively participating in the development of green finance disclosure standards. He also shared Hong Kong’s experience in issuing green tokenized bonds to stimulate the relevant market.
At the conclusion of the thematic forum, guests including Zhou Xiaochuan, former Governor of the People’s Bank of China, engaged in a discussion with the speakers on topics such as the impact of AI technology on the banking industry.
In response to questions from the audience, Lv Zhongtao added that the current role-driven approach stems from banks’ practical and objective operational needs. In contrast, the U.S. startup model features smaller tech companies with fewer employees, allowing for the incubation of entirely new scenarios and the simulation of novel processes while maintaining room for error. However, existing large banks, operating under strict regulatory oversight, require a gradual process to integrate AI. Currently, they are merely embedding AI tools into existing roles rather than using AI technology to fundamentally restructure their business models. He believes that, from a technological perspective, banks will continue to evolve from the current process-oriented model toward an AI-native business model. In the past, the structural setup of banks—divided into front, middle, and back offices—was primarily characterized by heterogeneous decentralization. In the future era of large language models, this heterogeneity may manifest as real-time monitoring where models supervise other models, or where heterogeneous models oversee other heterogeneous models, rather than the current workflow between distinct roles. Regarding individual capabilities and role boundaries, future model training will expand these boundaries, integrating capabilities across multiple roles to reconstruct existing business processes.
