The Financial Stability Board has opened a consultation on sound practices for the responsible adoption of artificial intelligence by financial institutions. The proposals set out 12 practices.
The measures are intended to guide firms on governance, risk management and operational controls as they adopt AI across their businesses. They cover organisation-wide oversight, the management of AI risks throughout development and deployment, and the handling of cyber, information and communication technology, and third-party risks associated with AI use.
The consultation focuses on AI-specific issues the FSB considers relevant to financial institutions and financial stability. It is aimed at boards and senior management, which should use the practices as a reference point when weighing business strategy, technology adoption and risk management.
The proposed framework does not create an international standard or prescribe how firms should adopt AI. It is also not designed to address recent risks tied to frontier AI models, although some of the practices could help institutions respond to those issues.
Three areas
The 12 practices are grouped into three broad areas. The first four deal with organisation-wide AI governance; the next six address the management and mitigation of AI risks during development and deployment; and the final two focus on cyber, technology, and third-party risks.
The package builds on earlier work by the FSB and other standard-setting bodies and is intended to be broadly compatible with existing regulatory and supervisory efforts. It also seeks to improve coordination, cooperation, and information sharing among financial institutions, supervisors, and other stakeholders across jurisdictions.
The move comes as regulators and market participants assess how wider AI use could affect the resilience of banks, insurers and other financial firms. The FSB has previously examined the financial stability implications of AI in the financial system and has updated that work as adoption has widened.
International policymakers have increasingly focused on whether firms have the internal governance and controls needed to use AI tools without creating new operational or conduct risks. Concerns include model errors, weak oversight of third-party providers, cyber vulnerabilities and the speed at which more advanced systems are being introduced into financial services.
Michelle Bowman, Chair of the FSB Standing Committee on Supervisory and Regulatory Cooperation and Vice Chair for Supervision of the Board of Governors of the Federal Reserve System of the United States of America, set out the organisation's position on the consultation.
"This report establishes clear safeguards for financial institutions to adopt, innovate, and use AI responsibly. The report reflects significant collaboration among FSB members on an accelerated timeframe to keep pace with the rapid changes from advancements in AI. I look forward to receiving public feedback on this report, so that a final report can be issued later this year as a US G20 deliverable," said Michelle Bowman, Chair of the FSB Standing Committee on Supervisory and Regulatory Cooperation and Vice Chair for Supervision of the Board of Governors of the Federal Reserve System of the United States of America.
Regulatory context
The FSB coordinates work among national financial authorities and international standard-setting bodies on regulatory, supervisory and other financial sector policies related to financial stability. Its members include authorities from 24 countries and jurisdictions, as well as international financial institutions, sector-specific groupings of regulators and supervisors, and committees of central bank experts.
It also conducts outreach with about 70 other jurisdictions through six Regional Consultative Groups, extending its reach well beyond its formal membership. That gives the consultation added weight for firms operating across borders, even though the proposals are not framed as binding rules.
Ho Hern Shin, Lead of the SRC Workstream on Artificial Intelligence and Deputy Managing Director of the Monetary Authority of Singapore, said the pace of change in AI was a central reason for issuing the practices now.
"The recent developments in frontier AI models highlight the dynamic nature of this technology and the rapid pace at which its capability evolves. The FSB's sound practices are designed to help financial institutions navigate their AI adoption responsibly in a rapidly changing technology landscape," said Shin.