Review IT, security, procurement and software development lifecycle processes so they interoperate with AI due diligence; promote diverse decision-making on AI risk; develop incident monitoring and response systems; develop or adapt complaint and whistleblower procedures for RBC issues; foster critical thinking in AI design, development, deployment and use; develop processes to upgrade, decommission and phase out AI systems safely; develop contingency plans for failures, incidents and adverse impacts; develop a stakeholder engagement plan that lets stakeholders, including workers and unions, assess and monitor implementation; and join collaborative initiatives on shared standards and tools for safe, secure and trustworthy AI.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.