Glossary term
Glossary term
Governance and Compliance
Ongoing collection and analysis of performance, incidents, feedback, and risk signals after deployment. It ensures compliance and controls remain valid in real-world operation. Monitoring should be risk-based and connected to incident response, model updates, customer feedback, vendor notifications, and regulatory reporting obligations.
EU AI Act Article 72 requires providers of high-risk AI to establish and document a post-market monitoring system that systematically collects performance data.
FDA's Predetermined Change Control Plans for AI medical devices and the 2023 Real-World Performance Monitoring guidance require ongoing performance surveillance.
Salesforce's Trusted AI process includes a post-deployment monitoring layer for generative features, tracking incidents and prompting human review when thresholds trigger.