Loading video player...
**How to Build Visibility, Control, and Trust Into AI Operations** In this video, I walk you through the dashboard I built using AI tools to monitor various aspects of our system in accordance with NIST and OWASP guidelines. I cover key sections including system health, agent logs, and alerts related to NIST guidelines, highlighting the importance of availability, confidentiality, and integrity. I also demonstrate how the dashboard captures prompt injections and configuration changes, providing insights into agent performance and system status. Please take a moment to review the alerts and outputs generated by the agents, as your feedback will be crucial for further improvements. * OWASP highlights prompt injection, excessive agency, insecure output handling, and data leakage as monitoring-relevant risk surfaces. * NIST emphasizes ongoing monitoring, governance, and measurement as core to trustworthy AI operations. * Enterprise architecture guidance consistently treats observability, alerting, and audit trails as foundational controls for production systems.