Ongoing monitoring
Continuing checks that a model or agent is still performing as intended as data, use and conditions change.
Ongoing monitoring is the continuing evaluation of whether a model is performing as intended after it goes into use, as data, products and conditions change. The US agencies' revised model risk guidance treats it as a core part of validation, alongside outcomes analysis, and expects its frequency and scope to fit the model's risk. For agents, monitoring covers action outcomes as well as outputs: payments made, limits reached, approvals requested and errors corrected.
Agent Minute explains this term on 23 November 2026.
Related terms
Model validationEvaluating whether a model performs as expected, including its reliability and its limitations.Model driftA decline in a model's or agent's performance over time as data, users or conditions change.Post-market monitoringA provider's ongoing collection and review of experience from a deployed AI system, to find and fix problems.Audit trailA chronological record of who accessed a system and what operations they performed, enough to reconstruct events.