IOSCO's AI work is non-binding guidance for securities regulators, in three steps. FR06/2021 (September 2021) set six measures for supervising AI and machine learning at market intermediaries and asset managers; CR/01/2025 (March 2025) surveyed use cases, risks and challenges in capital markets; and the Supervisory Toolkit for AI Use in Capital Markets, FR/02/2026 (May 25, 2026), gives supervisors practical tools across governance, third parties, disclosure and recordkeeping, covering machine learning, generative AI and agentic AI. Nothing binds a bank directly, but where a bank runs a securities business, national regulators are being given IOSCO's questions to ask.
| Full name | International Organization of Securities Commissions |
| Role | Global securities-markets standard-setter; non-binding AI guidance and supervisory tools for securities regulators |
| Force on banks | Non-binding standards |
| Applies to | Securities regulators that are IOSCO members, and through them market intermediaries, asset managers and other supervised capital-markets firms. Banks are in scope where they run broker-dealer, trading, advisory or asset-management businesses supervised by a securities regulator. |
| Key document | FR/02/2026, 'Supervisory Toolkit for AI Use in Capital Markets' (Final Report, May 25, 2026), building on FR06/2021 (September 2021) and CR/01/2025 (March 2025) |
| Latest move | On May 25, 2026 IOSCO published its final Supervisory Toolkit for AI Use in Capital Markets (FR/02/2026) with a standalone extract (OR/07/2026) for use in examinations and inspections; feedback from stakeholders was invited by June 26, 2026, and the next phase is a review of emerging industry practices |
| Documents tracked | 2 · all documents → |
IOSCO does not make law. Its Board publishes reports that member regulators may adopt, and its AI output has moved from measures for firms (FR06/2021, the 2021 report) to a supervisory toolkit (FR/02/2026). The 2026 final report says its toolkit is 'non-binding, non-prescriptive', built on the multi-phase approach of IOSCO's Fintech Task Force and its AI Working Group, and informed by a survey to which 21 members responded and by roundtables in Tokyo, Singapore and New York.
The toolkit has three layers: areas of supervisory consideration, tools for four key areas (governance and risk management; third-party and outsourcing risk management; disclosure; recordkeeping and reporting) and indicators for monitoring AI adoption. It also introduces risk-based and proportionate supervision, with a four-level classification of human oversight and an explicit statement that larger, systemically important institutions may face heightened supervisory expectations even for medium- or low-risk AI applications. IOSCO says its next phase is a review of emerging industry practices on disclosure, recordkeeping and reporting, and governance, and that it will keep coordinating with the Financial Stability Board.
What has the IOSCO actually published on AI?
| Date | Document | Status |
|---|---|---|
| May 25, 2026 | IOSCO FR/02/2026 — Supervisory Toolkit for AI Use in Capital Markets: Final Report | In force |
| Sep 7, 2021 | IOSCO FR06/2021 — The use of artificial intelligence and machine learning by market intermediaries and asset managers: Final Report | In force |
| Date | Type | Document / event |
|---|---|---|
| Jun 26, 2026 | Milestone | Stakeholder feedback on the AI supervisory toolkit closes. IOSCO invited feedback on the toolkit, and on emerging industry practices, through a short survey open until June 26, 2026; it said responses will inform the next phase of its AI work. |
| May 25, 2026 | Guidance | IOSCO FR/02/2026 — Supervisory Toolkit for AI Use in Capital Markets: Final Report. IOSCO's Supervisory Toolkit for AI Use in Capital Markets (Final Report FR/02/2026), published May 25, 2026, gives securities supervisors 'practical, non-binding, non-prescriptive supervisory tools' for overseeing AI used by regulated firms across the AI lifecycle, from traditional machine learning to generative and agentic AI. source ↗ |
| Mar 12, 2025 | Milestone | IOSCO publishes consultation report CR/01/2025 on AI in capital markets. 'Artificial Intelligence in Capital Markets: Use Cases, Risks, and Challenges' sought comment by April 11, 2025. The 2026 toolkit summarises its most-cited risk areas as malicious uses, model and data considerations, concentration, outsourcing and third-party dependency, and interactions between humans and AI. |
| Sep 7, 2021 | Guidance | IOSCO FR06/2021 — The use of artificial intelligence and machine learning by market intermediaries and asset managers: Final Report. IOSCO Final Report FR06/2021, published September 7, 2021 (media release IOSCO/MR/23/2021), gives six measures for regulators to consider, three of which (Measures 2, 3 and 4) regulators 'should require', for market intermediaries and asset managers using AI and machine learning. source ↗ |
- IOSCO's review of emerging industry practices on disclosure, recordkeeping and reporting, and governance of AI use, which the 2026 report names as the next phase
- National securities regulators adopting the toolkit's questions in examinations of bank-owned broker-dealers and asset managers
- Alignment with the Financial Stability Board's AI sound practices work, which IOSCO says it will continue to coordinate with
Is IOSCO's AI guidance binding on banks?
No. IOSCO reports are non-binding and addressed to securities regulators, which may adopt them in their own rules. They reach a bank only through national regulation of its securities, advisory or asset-management activities.
What is the IOSCO AI Supervisory Toolkit?
It is IOSCO's final report FR/02/2026, published May 25, 2026, giving supervisors tools to oversee AI used by regulated firms across the AI lifecycle, from traditional machine learning to generative and agentic AI. It sets out areas of supervisory consideration, tools for four key areas, and indicators for monitoring AI adoption.
How is the 2026 toolkit different from the 2021 IOSCO AI report?
FR06/2021 gave six measures for regulators to consider on AI and machine learning at market intermediaries and asset managers. The 2026 toolkit builds on those measures, extends to generative and agentic AI, and focuses on practical supervisory tools and example questions rather than high-level measures.
Follow every move these regulators make
when one of these regulators moves, the next morning's brief says so · six sourced stories · 7 am ET · free
plus every tracker, bank and agent page update, the morning after · leave any morning