What's new
Overview
This deep dive examines the latest developments in artificial intelligence, including emerging adoption patterns, key risk watch points, evolving policy and supervisory guidance, and research on the economic impact of AI.
What's new
Sweden's Financial Supervisory Authority selects reference group for Innovation Compass project on AI-based ESG data reliability
Sweden's Financial Supervisory Authority (Finansinspektionen, FI) has appointed industry participants to a reference group for its first Innovation Compass guidance project on the reliability of automated and AI-based methods for ESG data. Through workshops, the project will identify practical questions and challenges in using such data for external sustainability reporting and for assessing and managing sustainability risks, with participants including Climate Tracker Initiative, Metria AB, SEB, Handelsbanken, SBAB Bank, Norion Bank and Länsförsäkringar AB.
National Bank of Denmark analysis estimates AI could add 0.1–1.0 percentage points to Denmark’s annual total factor productivity growth
The National Bank of Denmark published an analysis concluding that artificial intelligence is likely to support higher productivity in the Danish economy, but with highly uncertain magnitude and timing, so no separate AI contribution is included in its latest three-year forecast. The analysis notes rapid AI adoption, with 42 per cent of Danish companies using at least one AI technology in 2025, and estimates that 39–61 per cent of work tasks are exposed to AI. It finds AI could raise annual total factor productivity growth by around 0.1–1.0 percentage points over ten years, compared with about 0.5 per cent over the past five years, and highlights potential second-round effects via investment, labour market reallocation, unit costs and energy demand from data centres.
Dutch Authority for the Financial Markets finds AI adoption rising in Dutch asset management while governance and ethics lag
The Dutch Authority for the Financial Markets has published research finding that artificial intelligence use in the Dutch asset management sector is increasing faster than firms’ policies and internal arrangements, with associated risks growing. Asset managers mainly use AI for analysis, price forecasting and trading strategies, but many lack budgets, AI-specific governance, policies on employee and generative AI use, and ethics frameworks. The authority urges firms to strengthen governance, including clear policies, data quality controls, supplier due diligence and explainable, controllable models.
State Bank of Vietnam flags AI-driven cyber risks and urges joint action to secure AI systems
The State Bank of Vietnam’s Deputy Governor Pham Tien Dung used a National Cybersecurity Association workshop in Hanoi to outline how artificial intelligence is reshaping banking operations and cyber risks, calling for closer coordination between authorities, businesses and experts. He highlighted AI applications in credit assessment, fraud detection, process automation and customer support, alongside emerging threats such as personalized scams, deepfake-enabled impersonation and adaptive malware, and referenced efforts to refine Vietnam’s cybersecurity and AI governance framework with a focus on safety, transparency and accountability.
Federal Reserve Board research reconciles AI adoption surveys showing 18% of US firms using AI by end-2025 and broad workforce exposure
The Federal Reserve Board published a research note comparing three major surveys on U.S. artificial intelligence adoption, finding that about 18% of firms reported AI use by end-2025, while around 41% of workers reported using work-related generative AI and 78% of the labor force is employed at AI-adopting firms. The note attributes differences in adoption rates mainly to survey design, including sampling, units of analysis, question framing, and respondent knowledge. Adoption increases with firm size and is notably high in professional services and finance.
China Banking and Insurance Regulatory Commission starts Jin Jian regtech programme and calls for orderly financial-sector AI pilot deployments
The China Banking and Insurance Regulatory Commission held an expanded Party committee study session on artificial intelligence to set priorities for AI use in financial services and supervision, positioning it as a strategic focus for technology finance and the sector’s digital transformation. Director Li Yunze called for building a financial service system aligned with full life-cycle financing needs of future industries, prudently piloting AI applications, establishing a secure AI governance framework, and launching the “Jin Jian Project” to strengthen supervisory technology and look-through supervision.
Union of Arab Securities Authorities reports AMERC annual meeting focus on IOSCO 2026 priorities and tackling online fraud
The Union of Arab Securities Authorities reported that the IOSCO Regional Committee for Africa and the Middle East held its 2026 annual meeting virtually, focusing on IOSCO’s 2026 work programme and regional activities. Participants reviewed regional market developments, capacity-building initiatives, and survey results on members’ priorities, and discussed measures to enhance market resilience, combat online fraud, strengthen investor protection, and address supervisory challenges including technology, artificial intelligence, and cross-border cooperation.
Croatian National Bank and European Investment Bank Group highlight low AI uptake and financing skills and policy barriers in Croatia
The Croatian National Bank reported on a joint conference with the European Investment Bank Group on building an “AI-ready” Croatian economy, where new EIB research showed that while 89% of Croatian firms are investing, only around one fifth systematically use generative artificial intelligence, below the European Union average in AI adoption. Speakers identified access to finance, skills shortages, fragmented data and regulatory uncertainty as key barriers, and highlighted financing, skills and policies as levers to accelerate AI use, while noting risks from overreliance on AI and labour market impacts.
European Central Bank research finds 38% of euro area firms are advanced AI adopters and users plan higher AI investment
The European Central Bank published analysis from its Survey on the Access to Finance of Enterprises for the fourth quarter of 2025, examining euro area firms’ artificial intelligence adoption and investment plans. The study finds uneven but substantial AI uptake, with advanced users more likely to anticipate higher turnover, fixed-asset investment, wages and employment, and planning a larger share of total investment in AI than non-users and early-stage adopters.
Monetary Authority of Singapore outlines insurer priorities with revised Notice 133 capital rules effective 31 March and planned guidance on AIR, third-party risk and AI
The Monetary Authority of Singapore outlined supervisory priorities for insurers around fair dealing and customer outcomes, risk-appropriate capital and risk management, operational resilience, and disciplined adoption of artificial intelligence. Key measures include classifying investment-linked policies as complex products, revising Notice 133 under the risk-based capital framework to adjust capital treatment for infrastructure, structured products and equities, and enhancing oversight of private asset exposures. MAS is also consulting on new Guidelines on Third Party Risk Management and Operational Risk Management and developing AI Risk Management Guidelines.
Federal Reserve Board research finds no evidence that higher AI adoption has reduced firms’ job postings
The Federal Reserve Board's FEDS Note finds no evidence that higher AI adoption levels correlate with reduced job postings, using data from September 2023 to November 2025. The analysis, combining Lightcast job postings and US Census Bureau data, shows industry-level effects are generally positive but not statistically significant, with firm-level results indicating precisely estimated null effects. The findings are framed as backward-looking and not causal, suggesting ongoing monitoring as AI adoption continues.
U.S. Financial Services Committee publishes oversight remarks urging regulators to keep pace with AI and digital-asset innovation
The U.S. Financial Services Committee released remarks and testimony from federal banking regulators on adapting supervisory frameworks to rapid innovation in AI and digital assets. Key discussions included the adequacy of current agency structures, the need for new supervisory technologies, and regulatory certainty across administrations. Criticism was directed at the Federal Reserve's Novel Activities Supervision Program, with calls for legislative changes in stablecoin and digital asset markets.
Bank for International Settlements' Financial Stability Institute reviews emerging supervisory approaches to generative AI data use in financial services
The Bank for International Settlements' Financial Stability Institute released a paper on data use and risks in financial institutions' AI applications, focusing on generative AI. It highlights challenges like fragmented legacy data and third-party dependencies, and suggests more tailored supervisory guidance on AI data governance, quality standards, and security. The paper also calls for enhanced collaboration between financial and data protection authorities to address cross-border issues.
U.S. Senate Committee on Banking, Housing and Urban Affairs ranking member Warren criticises Trump Administration deregulation in statement on FSOC meeting
Senator Elizabeth Warren warned of rising financial stability risks after a Financial Stability Oversight Council meeting, criticizing the Trump Administration's weakening of post-crisis oversight. She highlighted concerns over private credit deterioration, an oil price shock linked to the President's actions in Iran, and a potential debt bubble in the artificial intelligence sector. Warren argued that deregulation could lead to conditions necessitating taxpayer bailouts and broader economic harm.
Office of the Superintendent of Financial Institutions signals targeted capital recalibration and a 12-month fast-track approvals process for eligible new entrants
Superintendent Peter Routledge of the Office of the Superintendent of Financial Institutions (OSFI) outlined a strategic shift towards fine-tuning capital standards and regulatory processes, emphasizing evidence-based adjustments. OSFI is recalibrating capital requirements for SMEs and infrastructure investments, modernizing regulatory guidance, and developing a fast-track approvals framework for new market entrants. Supervisory priorities include monitoring AI-related governance risks, housing credit risks, and banks' exposures to private credit funds, focusing on insurers' private market exposures and alignment with IFRS 18.
Commodity Futures Trading Commission launches Innovation Task Force to shape regulatory framework for crypto, AI and prediction markets
The Commodity Futures Trading Commission has launched an Innovation Task Force to establish clearer regulations for novel products and technologies in U.S. derivatives markets. The task force, in collaboration with the Innovation Advisory Committee, will focus on crypto assets, blockchain technologies, artificial intelligence, and prediction markets. It will also coordinate with other federal agencies, including the U.S. Securities and Exchange Commission.
Financial Transactions and Reports Analysis Centre of Canada publishes FIFAI II report on AI risks and the AGILE framework for financial services
FINTRAC released the final report from the Financial Industry Forum on AI Phase II, highlighting AI-related risks and opportunities in Canadian financial services. The report introduces the AGILE framework, focusing on strategic risks, cybersecurity threats, consumer protection, and financial stability, with priorities including enhanced oversight, governance, and AI literacy. The findings reflect individual insights from forum participants and are not official guidance from FINTRAC or other Canadian authorities.
European Central Bank’s Lane highlights rapid euro area AI adoption and sets a 2026-27 plan to embed AI across ECB workflows
ECB Executive Board member Philip R. Lane discussed AI's uncertain impact on the euro area economy, productivity, employment, and inflation. The ECB aims for systematic AI implementation by 2027, amid rising AI adoption and digital investment lagging behind the US. Lane noted financing frictions for AI investment and risks for banks, while the ECB enhances its digital capabilities with machine-learning tools and a central banking digitalisation programme.
U.S. Department of the Treasury launches AI Innovation Series with four public-private roundtables on financial sector AI adoption
The U.S. Department of the Treasury launched the Artificial Intelligence Innovation Series, a public-private initiative to explore AI deployment and oversight in the U.S. financial system. The series will include four roundtables with stakeholders to identify AI use cases and approaches to scaling innovation while ensuring safety. Treasury Secretary Scott Bessent emphasized the need for evolving governance and regulatory frameworks to support AI in financial services.
Australian Securities & Investments Commission adds Moneysmart guidance on using publicly available AI tools for money questions
The Australian Securities & Investments Commission (ASIC) has issued new guidance on its Moneysmart website to assist Australians using AI tools for financial information. It addresses the growing reliance on AI among Gen Z, highlighting potential inaccuracies and advising verification against trusted sources. ASIC also announced an updated Moneysmart website, celebrating 15 years of providing free consumer education.
Monetary Authority of Singapore concludes Project MindForge phase two with an AI risk management toolkit for the financial sector
The Monetary Authority of Singapore concluded phase two of Project MindForge by releasing an AI Risk Management Toolkit for financial institutions. Developed with 24 industry partners, it includes an AI Risk Management Operationalisation Handbook and case studies, structured around MAS’ proposed Guidelines on AI Risk Management. MAS will periodically update the handbook and establish a workgroup under BuildFin.ai to enhance AI risk management resources and capabilities.
European Central Bank issues opinion on EU AI Act simplification proposal with bank supervision information sharing and credit scoring model carve outs
The European Central Bank (ECB) has issued an opinion on amending the Artificial Intelligence Act, advocating for clearer separation between bank supervision and AI market surveillance. The ECB suggests legal provisions for information sharing with national market surveillance authorities and calls for clarification on "high-risk" credit scoring techniques. It recommends excluding certain models from high-risk classification and involving other European financial authorities in guidance for regulated firms.
Sweden's Financial Supervisory Authority invites applications to join Innovationskompassen reference group on reliability of AI-based ESG data
Sweden's Financial Supervisory Authority seeks expressions of interest to join a reference group for its Innovationskompassen project, focusing on the reliability of automated and AI-based ESG data methods. The project aims to address supervisory guidance needs by mapping challenges in ESG data use, including sustainability reporting and risk management. Selected participants will engage in workshops from April to June 2026.
Bank of Italy releases six new Occasional Papers on artificial intelligence, climate data gaps, sustainable finance pricing and corporate lending
The Bank of Italy published six Occasional Papers (Nos. 1001–1006) examining AI's macroeconomic and firm-level effects, sustainable finance pricing, and corporate lending by Italian less significant institutions. Key topics include AI's impact on the US economy, effects on Italian firms, and the use of generative AI in policy briefing and climate data. The studies also explore environmental scores' influence on bond pricing and patterns in corporate lending.
European Council presidency convenes macroeconomic dialogue with social partners on fragile economic outlook and AI’s impact on EU labour markets
The European Council presidency, European Central Bank, European Commission, and Eurogroup President held a Macroeconomic Dialogue with European social partners on economic developments and AI's impact on the EU labour market. The dialogue highlighted the EU's improving yet fragile outlook amid geopolitical tensions and emphasized AI's dual role as a productivity driver and a risk to social cohesion. The Commission noted proposed changes in the Digital Omnibus to support responsible AI uptake, while social partners called for stronger worker protections and regulatory simplification.