Complete and retain pre-deployment impact assessments for high-risk AI systems addressing purpose, data inputs, training data, performance metrics, and risks to consumers.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.