Artificial Intelligence Governance, Risk Management, Compliance and Supervision in Capital Markets

- Register and complete the payment through the official training website.

- Upon successful registration, an invitation email—including the access link, workshop resources, and Pre- & Post-Assessment links—will be sent one day before the workshop.

Artificial Intelligence is moving rapidly from experimentation to practical deployment across financial services and capital markets, creating significant opportunities for efficiency, enhanced analytics, automation and improved decision-making. At the same time, it creates or amplifies risks relating to model performance, data quality, bias, cybersecurity, operational resilience, investor protection and accountability. This three-day executive programme examines AI through the specific lens of capital-market governance, risk management, compliance and supervision. Its purpose is to enable participants to make better governance, regulatory and supervisory decisions about AI. 

• Securities and capital-market regulators; stock exchanges and market institutions • Board members, CEOs, responsible officers and senior executives of regulated institutions • Chief Risk Officers, Chief Compliance Officers, General Counsel and legal departments • Internal audit, internal-control and market-surveillance professionals • Chief Data, Digital, Technology and AI Officers • Investment, asset-management, brokerage and securities-firm executives • Regulatory-policy and supervisory professionals • Professionals responsible for digital transformation, RegTech and SupTech • Policymakers involved in AI, financial regulation and national digital strategies 

Certificate of participation issued by UASA & CMA. 

DAY ONE — GOVERNING ARTIFICIAL INTELLIGENCE IN LICENSED CAPITAL-MARKET INSTITUTIONS

Session 1 — AI in Capital Markets: From Innovation Opportunity to Regulatory Responsibility

• The evolving role of AI in capital markets • Investment research, portfolio management, investment advice, brokerage and client interaction

• Trading and execution; compliance and financial-crime monitoring; market surveillance • Risk management and internal audit

• Generative AI and emerging agentic AI applications • Internal versus client-facing applications; identifying material AI use cases

• When AI becomes a governance and regulatory issue

Executive Question: When does an AI initiative stop being an IT project and become a Board and regulatory issue? 

Session 2 — Governing AI: Board Oversight, Senior Management and Human Accountability

• AI strategy, institutional governance and Board risk appetite

• Senior-management accountability and allocation of AI ownership

• AI governance committees and the roles of compliance, risk, legal, technology and internal audit

• Human oversight and meaningful human control

• Delegation to AI versus delegation of responsibility

• AI policies, approval processes, escalation, inventories and classification • Governance proportionality, materiality and Board reporting 

• Executive Question: What must a Board know before approving a material AI use case?

Executive Question: What must a Board know before approving a material AI use case?

Session 3 — AI Risk Management: From Model Risk to Enterprise Risk

• AI risk taxonomy; model performance and model risk • Data quality, data governance, bias and discriminatory outcomes

• Explainability, opacity, hallucinations and unreliable outputs

• Automation bias and human overreliance

• Cybersecurity, privacy, confidentiality and operational resilience

• Third-party dependencies and concentration risk

• Conduct and investor-protection risks

• AI risk registers, control mapping, lifecycle monitoring and reassessment

Executive Question: Which AI risks can be accepted, which must be controlled, and which should prevent deployment? 

Session 4 — Day One AI Governance Case Lab: “Should We Deploy It?”

• Board/senior-management assessment of an AI solution supporting investment recommendations

• Concerns: client and training data, explainability, suitability, bias, vendor dependence and hallucinations

• Human review, compliance approval, client disclosure and ongoing monitoring

• Decision: APPROVE ? APPROVE WITH CONDITIONS ? PILOT ? ESCALATE ? REJECT

• Justification through Governance ? Risk ? Compliance ? Investor Protection ? Regulatory Expectations

DAY TWO — SUPERVISING ARTIFICIAL INTELLIGENCE IN CAPITAL MARKETS

Executive Recap (15–20 minutes): Key lessons from Day One, participant questions, reconsideration of the deployment decision and transition from the institution’s perspective to the supervisor’s perspective.

Session 5 — International Standards and Principles for Responsible AI in Financial Markets • Emerging international architecture for responsible AI 

• Risk-based and proportionate governance; accountability and human oversight 

• Transparency, explainability, robustness, security and resilience

• Data governance and model lifecycle governance

• Third-party dependencies; investor and consumer protection

• Financial-stability considerations

• Translating international principles into capital-market supervision

• Comparative reference to IOSCO, FSB, OECD, ISO and selected financial-sector governance approaches

Session 6 — Comparative Regulatory Approaches: International, European & MENA Perspectives

• National AI strategies versus financial-sector regulation

• Horizontal AI regulation versus sector-specific financial regulation

• Principles-based versus prescriptive approaches; application of existing securities regulation to AI

• European Union: EU AI regulatory architecture and ESMA supervisory perspectives

• France: AMF market and supervisory perspectives; HCJP legal and regulatory analysis

• United Kingdom: FCA approach, AI testing and regulatory experimentation

• Arab & MENA jurisdictions: selected securities-regulatory developments, supervisory approaches and national AI strategies

• Regulatory sandboxes and controlled experimentation

• Lessons that can—and cannot—be transplanted across jurisdictions

Executive Question: Which international regulatory approaches can realistically be adapted to Arab capital markets, and which require adjustment?

Session 7 — Supervising AI: What Should Securities Regulators Examine?

• Developing a risk-based supervisory approach and mapping AI adoption

• Identifying material and higher-risk use cases; AI inventories and model registers

• Governance documentation and Board/senior-management oversight

• Model validation, independent challenge and data governance

• Third-party AI providers and human oversight • Monitoring, incident reporting and regulatory reporting

• Supervisory interviews; on-site and off-site supervision

• Testing governance effectiveness; supervisory capacity and skills

• When should a supervisor intervene?

Executive Question: What evidence would convince a regulator that an institution genuinely controls its AI rather than merely having an AI policy?

Session 8 — Day Two Supervisory Inspection Lab: “The AI-Enabled Investment Firm”

• Participants act as a securities regulatory inspection team

• Review of a fictional AI policy, inventory, risk assessment & vendor documentation 

• Assessment of Board minutes, model-validation report, compliance assessment, incident log and client complaints

• Determine what is satisfactory, what is missing and what requires challenge

• Identify additional evidence, remediation and whether formal supervisory intervention is necessary

DAY THREE — AI COMPLIANCE, ACCOUNTABILITY, INCIDENTS AND REGULATORY RESPONSE

Executive Recap (15–20 minutes): Lessons from the supervisory inspection, key governance deficiencies, participant questions and transition from supervision to accountability and enforcement.

Session 9 — AI and Existing Capital-Market Regulatory Obligations

• Clients’ best interests; suitability and appropriateness; fair treatment of investors

• Disclosure, transparency and conflicts of interest

• Governance and systems-and-controls requirements

• Outsourcing and third-party arrangements

• Record keeping, data and confidentiality

• Market integrity and regulatory reporting

• Compliance monitoring, documentation and auditability

• Existing regulation versus AI-specific regulation; identifying potential regulatory gaps

Executive Question: Do capital markets need AI-specific regulation—or better application of existing regulatory principles?

Session 10 — AI Accountability: Who Is Responsible When AI Gets It Wrong?

• Institutional accountability and Board/senior-management responsibility

• Responsibilities of business, compliance, risk, technology and AI functions

• Third-party vendors and contractual allocation of responsibility

• Human-in-the-loop failures and overreliance on automated outputs

• Investor losses; civil and regulatory liability; administrative enforcement

• Cross-border AI systems and complex AI supply chains

Executive Question: If nobody intended the harmful outcome, who should nevertheless be accountable?

Session 11 — AI Incidents, Regulatory Investigations and Enforcement

• What constitutes a material AI incident; detecting failures and immediate containment

• Investor protection, human override and system suspension

• Evidence preservation, AI logs and audit trails • Internal investigation and root-cause analysis 

• Vendor involvement, regulatory notification and cooperation with supervisors

• Remediation, enforcement considerations, Board response and governance reform

• Illustrations: biased recommendations, hallucinated information, unsuitable advice, data leakage, model drift and third-party failure

Session 12 — Capstone AI Governance Crisis Simulation: From Incident to Regulatory Response

• Participants assume the roles of Board, Senior Management, CRO, CCO, General Counsel, AI/Technology Function and Securities Regulator

• Scenario: questionable AI recommendations, investor losses, complaints, deficient approval, vendor opacity and prior management warnings

• Decide whether to suspend the system and what immediate investor-protection measures are required

• Determine notifications, evidence preservation, governance adequacy and allocation of responsibility

• Assess supervisory/enforcement action, vendor responsibility, remediation and conditions for redeployment

• Final output: AI Governance & Supervisory Action Plan — Governance ? Risk ? Compliance ? Supervision ? Incident Response ? Accountability