Hi SupTech Community,
Welcome to the 30th issue of the Cambridge SupTech Lab bi-weekly LinkedIn newsletter, a source for updates on recent innovations, breakthroughs, opportunities, upcoming events in the suptech field, and news from the Cambridge SupTech Lab team.
If you would like to flag any items for inclusion in the next newsletter, please email us at cambridgesuptechlab@jbs.cam.ac.uk. Occasionally, we email our global community of supervisors, data scientists, vendors, and suptech experts to share event invites, news, or new courses—subscribe here.
This edition includes news from the Asian Development Bank, Alliance for Financial Inclusion, Bank for International Settlements, Central Banking, CGAP, European Central Bank , Global Government Fintech, International Association of Insurance Supervisors, International Monetary Fund, National Bank of Cambodia, Regtech Africa, Reserve Bank of India, and others.
Suptech Innovations
Reserve Bank of India (RBI) unveils MuleHunter.AI to combat fraud. The RBI Innovation Hub has introduced MuleHunter.AI, an artificial intelligence (AI)-powered tool to detect and flag mule accounts, helping reduce financial fraud and money laundering. Unlike traditional rule-based systems, MuleHunter.AI uses machine learning algorithms to analyse transaction data more accurately and quickly, enabling banks to identify fraudulent accounts faster. This initiative supports the RBI’s efforts to combat digital fraud, which accounted for nearly 68% of cybercrime complaints in Q2 2022 in India. Read more here.
The Bank Supervision Application (BSA) is transforming financial supervision across 21 supervisors, as highlighted in a new case study by the Alliance for Financial Inclusion (AFI). Used by 19 central banks and two financial authorities, the web-based platform streamlines regulation and monitoring, covering licensing, data collection, validation, reporting, risk management, and consumer protection. Future plans include integrating AI and cloud computing to enhance its capabilities. Read more here.
Central Banking & Technology
Machine learning can improve anomaly detection in Malaysia’s money services business (MSB) outlets by combining transactional and geolocation data, allowing for more efficient supervision – Bank for International Settlements (BIS) paper. Since 2017, licensed MSB operators have submitted transactional data to the Central Bank of Malaysia, allowing for off-site monitoring, but supervisors struggle with increasing data volumes and identifying high-risk patterns at the outlet level. The paper proposes a weakly-supervised machine learning approach that integrates transactional and geolocation data to enhance supervision and resource allocation. Read more here.
Climate fintech can drive investment in climate finance, but its full potential requires a broader policy approach addressing data access, infrastructure, regulation, and risk management. Climate fintech — the intersection of climate change, financial services, and digital technologies — plays a vital role in attracting investment to climate finance by leveraging innovative technologies such as blockchain, AI/ML, Internet of Things, geospatial technologies and digital twins to overcome financing barriers. While technology alone cannot fully solve these challenges, it offers promising solutions that address critical gaps. This note by the IMF staff explores the strengths and limitations of fintech in climate finance, highlights the private sector’s role, and discusses key lessons and policy implications for scaling up climate investments effectively. Read more here.
New data science project uses synthetic data to enhance money laundering detection. The Alan Turing Institute, Plenitude Consulting, Napier AI, and the Financial Conduct Authority (FCA) have launched a project to improve money laundering detection by creating a synthetic dataset using anonymised financial transactions. The dataset, developed to address the lack of realistic data for testing new detection methods, will be available through the FCA’s digital sandbox. It allows firms to test innovative money laundering detection techniques in a controlled environment. This initiative supports the FCA’s goal of advancing financial crime detection and increasing trust in financial services. Read more here.
Hong Kong Monetary Authority (HKMA) announces the inaugural cohort of the generative AI sandbox. The HKMA has unveiled the first batch of projects for its new ‘Generative AI Sandbox,’ aimed at fostering the responsible use of generative AI in the banking sector. Following the announcement in August 2024, 15 use cases from 10 banks and four technology firms were selected out of over 40 proposals. These projects primarily focus on enhancing risk management, anti-fraud measures, and improving customer experience, marking a significant step in the HKMA’s ongoing efforts to explore innovative AI applications in financial services. Read more here.
European Union’s Markets in Crypto-Assets (MiCA) crypto regulation now fully enforced. The EU’s MiCA regulation, officially coming into full force on December 30, sets the global standard for comprehensive crypto oversight. Approved by the European Parliament in April 2024, MiCA establishes a unified regulatory framework across the 27-member bloc, covering crypto-asset issuance, services, and platforms, including Bitcoin and other digital tokens. This landmark regulation aims to enhance transparency and safety within the crypto sector, with provisions for licensing, supervision, and operational standards for crypto firms. MiCA’s implementation is a significant step toward harmonising European digital finance regulation. Read more here.
Financial Stability Institute insights on Regulating AI in the financial sector: recent developments and main challenges. The paper explores the transformative potential of AI in the financial sector, focusing on operational efficiency, risk management, and customer experience in banking and insurance. It highlights the widespread adoption of AI, including generative AI, and its regulatory challenges, such as exacerbating risks like model risk and data privacy. While most financial regulations already cover these risks, areas such as data governance, expertise, model risk management, non-traditional players in the financial sector, new business models, and the role of third-party AI service providers require further regulatory attention. Read more here.
Strengthening Georgia’s cyber risk regulation, supervision, and testing is crucial for improving the National Bank of Georgia’s (NBG) preparedness, with key gaps identified in supervision, information sharing, and cyber testing practices – International Monetary Fund (IMF) report. The IMF’s technical mission to Georgia aimed to enhance the NBG cyber risk regulation, supervision, and testing. The mission assessed NBG’s current cyber risk frameworks and found that while regulations exist, significant gaps remain in supervision, information sharing, and cyber testing. The mission recommends developing a comprehensive cyber strategy for the authority to address these shortcomings and improve the financial sector’s resilience. Read more here.
Call for Papers: Bank Regulation and Supervision Practices in Southeast and East Asia. The Asian Development Bank Institute (ADBI) invites researchers to submit country-specific papers on the evolving bank regulatory and supervisory frameworks in Southeast and East Asia. The focus should be on key regulatory areas such as deposit insurance, macroprudential regulations, climate change, digitalisation, and governance. An upcoming conference, scheduled for March-April 2025, by the ADB will gather central bankers, think tanks, government officials, and academic experts to discuss these practices. Submissions (up to 3,000 words in English) are due by January 10, 2025, with preference given to full manuscripts. Read more here.
New research reveals that Word2Vec embeddings of central bank communications can enhance inflation forecasting. A recent study by Douglas Araujo, Nikola Bokan, Fabio Comazzi and Michele Lenza found that analysing the European Central Bank’s press conference statements using Word2Vec embeddings — numerical representations of words used in natural language processing — enhances core inflation predictions. Unlike traditional methods, Word2Vec outperformed sentiment and dictionary-based metrics, providing valuable insights even when factoring in the bank’s own inflation forecasts. Read more here.
Can AI revolutionise financial consumer protection? While financial authorities need to harness the power of AI to improve consumer protection, they must also address the novel risks and challenges that AI brings, such as new forms of financial fraud, cyber threats, and data privacy concerns. This podcast episode explores how the Central Bank of the Philippines is adopting AI-driven suptech through its chatbot, BOB, to strengthen financial consumer protection. Experts from the Central Bank of the Philippines, the Cambridge SupTech Lab, and CGAP share insights into this groundbreaking technology’s potential benefits and challenges. Listen more here.
The choice between implementing a retail central bank digital currency (CBDC), fast payment systems (FPS), or both depends on a country’s market features, infrastructure maturity, and policy goals, with both systems potentially complementing each other – BIS paper. Retail CBDCs and FPS enable instant transactions, but CBDCs represent central bank money, while FPS uses private money. Central banks view them as complementary tools to achieve policy goals like payment efficiency, competition, and financial inclusion, with decisions to choose either or both shaped by the country’s infrastructure and market conditions. Read more here.
Cambodia approves banks to handle crypto transactions. The National Bank of Cambodia (NBC) has unveiled a regulatory framework for crypto assets, permitting commercial banks and payment service providers to offer limited services involving backed cryptocurrencies and stablecoins. This new framework, signed by NBC Governor Chea Serey on December 26, aims to regulate transactions and business activities related to crypto assets. Read more here.
Costa Rica’s SINPE Móvil sees widespread adoption, boosting bank efficiency. Since its launch in 2015, Costa Rica’s retail fast payment system SINPE Móvil has gained significant traction, with 76% of the population over 15 years old actively using the platform as of 2024. The system, which allows individuals and businesses to make quick payments, has decreased cash withdrawals, with users averaging 180 transactions annually. SINPE Móvil’s rapid adoption has not only facilitated daily transactions, such as shopping and tipping but has also contributed to lower non-interest expenses for Costa Rican banks, improving efficiency. The success of SINPE Móvil is a prime example of how digital payment systems can transform economies and banking operations. Read more here.
Technology in the financial system can be a source of good, but it must be managed effectively to balance innovation with risk mitigation. In her remarks at the Fintech Ireland Summit 2024, Sharon Donnery, Deputy Governor of the Central Bank of Ireland, discussed the transformative potential of technology in the financial system, emphasising that while innovation can improve outcomes for consumers and the economy, it also introduces risks. She highlighted the role of regulators in ensuring that technology is deployed responsibly, focusing on safeguarding financial stability, consumer protection, and system integrity. Donnery stressed that the key challenge is not the technology itself, but how it is used and how risks are managed. Read more here.
National Institute of Standards and Technology (NIST) report highlights risks of synthetic content and solutions for digital transparency. A new NIST publication examines the impact of generative AI technologies, which can create realistic digital content but also raise concerns about trust and safety. It explores methods for authenticating, tracking, and detecting synthetic content, while addressing risks like harmful material and non-consensual imagery. The report emphasises digital content transparency as a key solution, though it cautions against false trust perceptions, and advocates for science-backed standards to promote responsible AI use and improve public trust in digital information. Read more here.
The Government AI Readiness Index 2024 highlights global progress in AI adoption. The Index evaluates AI preparedness of 188 countries, offering valuable insights for policymakers as they navigate the growing role of AI in public service delivery. It assesses 40 indicators across three key pillars: Government, Technology Sector, and Data & Infrastructure. It shows increasing adoption of AI strategies, especially in low- and middle-income countries, with notable progress from nations such as Ukraine, Costa Rica, Moldova, and Uzbekistan. The US leads globally, with strong performance from countries like Singapore, South Korea, and France. Read more here.
Events
BIS publishes overview of 2025 programme for financial sector supervisors. Over the year, the BIS will be hosting several events, including virtual seminars, webinars, conferences, high-level meetings, online courses and virtual meetings for financial sector supervisors. The programmes will touch on various themes including suptech, prudential supervision, financial inclusion, climate risks, insurance supervision, securities trading, cryptoassets, among others. Read more here.
Open finance offers great potential for financial inclusion but also presents significant risks regarding data security. At a recent workshop by the Financial Stability Institute, The World Bank, and CGAP, experts from 50 countries discussed the challenges and opportunities of open finance for financial inclusion and discussed considerations from a recently published paper. A consumer and data protection panel emphasised the need for a robust regulatory framework, consumer-first policies, and collaborative efforts to ensure the benefits of open finance while safeguarding consumers’ rights and data. Read more here.
