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
Bank of Canada assesses AI’s potential effects on productivity, employment and inflation
The Bank of Canada says AI could raise productivity and living standards, but the timing, scale and distribution of the gains remain uncertain. AI may reshape tasks and disrupt some jobs, while its limitations mean human oversight remains necessary. The Bank will monitor the transition while aiming to keep inflation at or around 2%.
Bank for International Settlements’ Fernando Restoy calls for broader bank supervision covering AI governance and resilience
Fernando Restoy, Chair of the Bank for International Settlements’ Financial Stability Institute, called for bank supervision to extend beyond AI governance to operational and strategic resilience. Supervisors should update model risk management expectations, test faster cyber response and recovery, and assess how AI-driven economic change could affect borrowers, portfolios and bank business models.
Hong Kong Securities and Futures Commission Chairman Kelvin Wong calls on boards to establish accountable AI governance
Hong Kong Securities and Futures Commission Chairman Dr Kelvin Wong called on boards to retain human accountability for AI and treat its use as a governance duty rather than an IT project. Directors should build AI literacy, establish firmwide controls and ownership, and scrutinize investment returns, risks, performance and incidents. He highlighted a wide gap between issuers’ AI disclosures and the adoption of structured governance frameworks.
National Bank of Denmark working paper finds AI adopters' employment growth 11% lower by late 2025
A National Bank of Denmark working paper finds that 2023 AI adopters recorded employment growth 11% below trend relative to non-adopters by late 2025. The shortfall was concentrated in firms with fewer than 100 employees and resulted from lower hiring, particularly in AI-exposed occupations. Economy-wide employment and wages remained relatively unaffected.
National Bank of Denmark finds AI adoption is slowing firm level employment growth without reducing total employment
The National Bank of Denmark finds that AI adopting firms have reduced employment growth mainly through fewer hires, particularly among younger and more highly educated workers. Total employment and wages show no clear effects, suggesting displaced hiring has been absorbed elsewhere. The central bank’s baseline projection assumes no aggregate employment impact over the next two to three years, although risks run in both directions.
Organisation for Economic Co-operation and Development finds agentic AI entering operational workflows as governance gaps persist
The Organisation for Economic Co-operation and Development finds that agentic AI is entering operational workflows, particularly for structured tasks with verifiable outcomes and bounded risks. Organisations are using limited autonomy, human approval checkpoints and layered technical safeguards. Evaluation, cybersecurity, traceability and cross-organisational accountability remain underdeveloped.
European Central Bank President Christine Lagarde calls for deeper capital markets to finance sovereign AI capacity
European Central Bank President Christine Lagarde called for deeper European capital markets to finance AI computing capacity, open models and access to frontier technology. She said rapid AI adoption could lift productivity by up to 4% over a decade, while closing Europe’s data center gap could cost as much as EUR 600 billion. The speech positions AI as a central purpose for the savings and investments union and warns supervisors to prepare for possible spillovers from a correction in US technology markets.
Monetary Authority of Singapore urges stronger AI governance and cyber defenses as financial sector adoption scales
The Monetary Authority of Singapore urged financial institutions to strengthen AI governance and cyber defenses as adoption expands beyond pilots. Pathfin.ai now has more than 300 participants, while cross-bank AI testing for scam detection is expected to produce findings by the end of 2026. MAS also warned of a sixfold rise in high-severity vulnerabilities and called for stronger multilayered defenses supported by AI.
European Securities and Markets Authority identifies rising correction and cyber risks despite resilient markets
The European Securities and Markets Authority warned that elevated technology valuations and strong investor optimism are increasingly detached from weaker economic conditions and geopolitical risks, raising the prospect of an abrupt market correction. Operational risks are also rising as frontier artificial intelligence intensifies cyber threats, while growing U.S. equity concentration leaves EU investors more exposed to a repricing.
UAE Cabinet reviews second phase Agentic AI strategy and FedAI government ecosystem
The UAE Cabinet’s Ministerial Council for Artificial Intelligence and Development reviewed the second phase strategy for expanding agentic AI across the federal government. Planned initiatives include the secure FedAI technical ecosystem, a customer experience lab and government projects to be launched in the next phase.
Bank for International Settlements assesses frontier AI cyber threats and finds authorities reinforcing existing resilience frameworks
A Bank for International Settlements paper finds that frontier AI is accelerating cyber attacks by automating vulnerability discovery, exploitation and complex operations. Financial authorities are mainly reinforcing existing operational resilience frameworks, with greater emphasis on faster patching, stronger governance, recovery capabilities and oversight of critical third-party dependencies.
South Korea's Ministry of Economy and Finance sets four AI cooperation priorities for the Knowledge Sharing Program
The Ministry of Economy and Finance set four KSP cooperation priorities covering AI ecosystems, sector and energy transformation, specialized skills and international coordination. It also plans to focus projects on four strategic fields and connect policy advice more directly with development finance, multilateral bank projects and private investment.
Hong Kong Securities and Futures Commission considers agentic AI guidance and outlines offshore renminbi market expansion
Hong Kong Securities and Futures Commission Chief Executive Officer Julia Leung said the regulator is considering guidance to clarify governance and human oversight expectations for agentic AI. She also outlined measures to deepen Hong Kong’s offshore renminbi market, including southbound Stock Connect readiness for renminbi counters, a fixed income and currency trading platform, and wider collateral use by the end of 2026.
Bank of France finds generative AI use at 67% of French firms with at least 20 employees but limited productivity gains
The Bank of France found that 67% of French firms with at least 20 employees used generative AI in early 2026, mainly experimentally or in support functions. Productivity gains remain limited, though 60% of firms using or planning to use AI expect gains over the next three years. Firms also anticipate continued investment, modest employment reductions and no significant fall in selling prices.
South Korea Financial Services Commission expands AI security testing eligibility to 75 firms
The South Korea Financial Services Commission expanded eligibility for its second frontier AI security test from 49 to 75 firms and will select up to 15 participants for one-year relief from network-separation rules. Selected firms must adopt alternative security controls and report their findings, while an interim review of the first round found that AI could scan large codebases rapidly and consistently for vulnerabilities.
Swedish Financial Supervisory Authority launches assessment of financial firms’ AI use and risk management
The Swedish Financial Supervisory Authority is assessing financial firms’ use and management of AI, including the impact of the EU AI Act. A targeted survey of a broad range of supervised firms will address emerging uses and risks, including operational resilience and cybersecurity.
Australia’s Council of Financial Regulators finds system resilient, backs unchanged macroprudential settings amid elevated risks
The Council of Financial Regulators maintained that Australia’s financial system remains resilient despite elevated geopolitical, market and cyber risks, and backed unchanged macroprudential settings. Borrower stress remains limited, but authorities emphasized sound lending standards and stronger operational resilience. A payments cyber crisis coordination framework will be published shortly, alongside ongoing work on critical third-party and AI-enabled cyber risks.
Bank of Korea finds AI-exposed industries accounted for 94% of youth employment decline
The Bank of Korea found that high-AI-exposure industries accounted for 268,000, or 94%, of Korea’s 285,000 decline in youth employment between June 2022 and June 2026. Losses were greater where AI was used for automation, while unemployment and employment exits also rose among highly exposed young workers. The note recommends apprenticeships, stronger support for workplace training and a better balance between technology and human capital incentives.
UAE Cabinet launches 32-advisor agentic AI system, adopts autonomous vehicle rules and federal financial sustainability framework
The UAE Cabinet launched an agentic AI system with 32 specialized advisors to assess policies, legislation and government initiatives. It also adopted autonomous vehicle rules and a framework embedding sustainability and environmental, social and governance considerations in federal financial policies. Additional measures cover cloud migration, AI education and international agreements on investment, digital policy and information exchange.
Eastern Caribbean Central Bank concludes 2026 AI summer camp with 18 client-focused projects
The Eastern Caribbean Central Bank concluded its 2026 AI and Python summer camp, where participants presented 18 projects addressing regional business and public sector needs. Six winning teams shared XCD 32,000, and the program shifted to externally recognized certification.
South Korea's Ministry of Economy and Finance outlines computing, manufacturing model and regional support for SME AI adoption
South Korea's Ministry of Economy and Finance outlined support for SME adoption of AI agents and physical AI. Measures include access to computing infrastructure, a manufacturing foundation model, standardized data-sharing arrangements and region-specific AI deployment support.
Saudi Arabia Insurance Authority signs HUMAIN agreement to advance insurance sector AI adoption
The Saudi Arabia Insurance Authority and HUMAIN have agreed to support AI and cognitive data center adoption across the insurance sector. The collaboration will explore priority use cases and contribute to a sector-wide strategy covering data quality, analytics, secure data sharing and governance.
Central Bank of Bahrain holds financial sector cybersecurity event focused on AI-era data risks
The Central Bank of Bahrain and the National Cyber Security Center held a cybersecurity awareness event for financial services professionals focused on data protection in the AI era. Sessions covered human risk, secure practices, artificial intelligence and emerging cyber threats.
Financial Stability Board chair warns of cyber risks from frontier AI models
Financial Stability Board Chair Andrew Bailey warned that frontier AI models could intensify cyber risk and called for safe model deployment, robust financial-sector recovery capabilities and resilient critical technology providers. He also cautioned that rising leverage, stretched valuations, market concentration and AI-related optimism could amplify a market correction.
European Central Bank assesses rapid growth in US tech borrowing and limited euro area bond market spillovers
A European Central Bank blog finds that growing US hyperscaler issuance has broadened the euro area corporate bond market without materially crowding out other issuers so far. Hyperscalers have about EUR 40 billion outstanding and account for nearly 10% of new euro-denominated non-financial corporate bond issuance. Further growth could increase concentration and borrowing costs as investors adjust portfolios and demand higher risk premiums.