In gathering vulnerability information from acquirers, users and public sources and investigating credible reports, the organisation logs, monitors and analyses all model inputs and outputs for possible security and performance issues; makes the users of its models (AI system producers and acquirers) aware of how to report potential security and performance issues; and monitors vulnerability and incident databases for AI-related concerns, including those affecting the ML frameworks and libraries its models are built with.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.