Embed responsible AI into oversight bodies, structures and processes: assign oversight of AI due diligence to senior management and broader AI RBC responsibility to the board; assign implementation responsibilities to the staff most able to raise or lower risk (development, data gathering and annotation, content moderation, design, procurement, consequential decision systems) with documented roles and lines of communication; build record-keeping to inventory AI systems and document their risks, take in feedback from other units, feed feedback into design, and map and communicate risks across business relationships; open channels between senior management and implementers; communicate policies to staff; train staff on their risk duties; and align units through cross-functional committees (possibly with independent external experts) and incentives linked to RBC objectives.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.