14
Ethical review boards should be established early in the development process to guide decision-
making and ensure responsible AI usage. These boards play a critical role in evaluating
potential ethical concerns such as bias, privacy, and societal impact, and help ensure that AI
Systems align with the firm’s values and regulatory obligations. Early involvement of ethical
oversight supports transparency, accountability, and trust in AI-driven operations.
Companies should maintain a catalog of laws, regulations, and guiding documents applicable
to their AI usage. In addition to AI-specific regulations and frameworks, existing market
regulations and data protection rules may also be impacted as governing bodies promote new
legislation for AI development and usage. Risk should work with Compliance and Legal
functions to understand how the evolving regulatory landscape will impact the company’s
approach to AI development, deployment, usage, and risk monitoring.
Best Practices for Regulatory Alignment
• Embedded compliance controls to prevent unethical trading behaviors (e.g., spoofing, wash
trades)
• Regulatory development monitoring
• Privacy protections, bias mitigation, and environmental impact assessments
• Regular audits with comprehensive documentation
• Ongoing communication with internal and external counsel when using market- or
customer-facing AI Tools
6. Organizational Readiness
Effective AI governance requires organizational readiness, starting with skilled cross-functional
teams, clearly defined roles and responsibilities, and structured onboarding for AI users. The
CCRO recognizes the importance of trader certification, segregation of duties, and ongoing
training to ensure responsible AI usage. The CCRO supports the development of
organizational capabilities that enable effective challenge and oversight of AI Systems.
6.1 Talent and Training
Organizational readiness begins with building skilled, cross-functional teams capable of
supporting the full AI lifecycle. Organizations which experiment with AI without sufficient
discipline or expertise can put themselves at significant risk. Firms should implement
structured onboarding programs that introduce users to AI governance principles, model
limitations, and ethical considerations. Certification programs can help define permissible AI
usage and reinforce accountability. Ongoing training is essential to keep pace with evolving
technologies, regulatory expectations, and emerging risks. By investing in talent development,
firms can ensure that AI Systems are deployed responsibly and aligned with strategic
objectives.
Training should cover technical skills (e.g., AI/ML fundamentals, statistics, relevant tools),
risk and governance (e.g., AI risk categories, governance framework, validation methods),
ethics and compliance (e.g., bias, fairness, regulatory requirements), and domain knowledge
(e.g., commodity markets, trading strategies, operational processes). Training depth should
vary by role: developers require deep technical training, validators need validation
methodology expertise, and users need operational training on limitations and proper usage.
Ethical review boards should be established early in the development process to guide decision-
making and ensure responsible AI usage. These boards play a critical role in evaluating
potential ethical concerns such as bias, privacy, and societal impact, and help ensure that AI
Systems align with the firm’s values and regulatory obligations. Early involvement of ethical
oversight supports transparency, accountability, and trust in AI-driven operations.
Companies should maintain a catalog of laws, regulations, and guiding documents applicable
to their AI usage. In addition to AI-specific regulations and frameworks, existing market
regulations and data protection rules may also be impacted as governing bodies promote new
legislation for AI development and usage. Risk should work with Compliance and Legal
functions to understand how the evolving regulatory landscape will impact the company’s
approach to AI development, deployment, usage, and risk monitoring.
Best Practices for Regulatory Alignment
• Embedded compliance controls to prevent unethical trading behaviors (e.g., spoofing, wash
trades)
• Regulatory development monitoring
• Privacy protections, bias mitigation, and environmental impact assessments
• Regular audits with comprehensive documentation
• Ongoing communication with internal and external counsel when using market- or
customer-facing AI Tools
6. Organizational Readiness
Effective AI governance requires organizational readiness, starting with skilled cross-functional
teams, clearly defined roles and responsibilities, and structured onboarding for AI users. The
CCRO recognizes the importance of trader certification, segregation of duties, and ongoing
training to ensure responsible AI usage. The CCRO supports the development of
organizational capabilities that enable effective challenge and oversight of AI Systems.
6.1 Talent and Training
Organizational readiness begins with building skilled, cross-functional teams capable of
supporting the full AI lifecycle. Organizations which experiment with AI without sufficient
discipline or expertise can put themselves at significant risk. Firms should implement
structured onboarding programs that introduce users to AI governance principles, model
limitations, and ethical considerations. Certification programs can help define permissible AI
usage and reinforce accountability. Ongoing training is essential to keep pace with evolving
technologies, regulatory expectations, and emerging risks. By investing in talent development,
firms can ensure that AI Systems are deployed responsibly and aligned with strategic
objectives.
Training should cover technical skills (e.g., AI/ML fundamentals, statistics, relevant tools),
risk and governance (e.g., AI risk categories, governance framework, validation methods),
ethics and compliance (e.g., bias, fairness, regulatory requirements), and domain knowledge
(e.g., commodity markets, trading strategies, operational processes). Training depth should
vary by role: developers require deep technical training, validators need validation
methodology expertise, and users need operational training on limitations and proper usage.
