Post-Market Monitoring
Ongoing surveillance of an AI system's performance, safety, and compliance after it has been deployed to production. Required under the EU AI Act for high-risk systems, post-market monitoring ensures that AI systems continue to meet their intended specifications as real-world conditions change.
Why It Matters
AI systems degrade. Data distributions shift, user behavior changes, and the world moves on from the conditions the model was trained on. Post-market monitoring catches these drifts before they become incidents.
Example
A medical device company monitors its AI diagnostic tool in production, tracking accuracy rates by patient demographic, flagging any drop in sensitivity below 95%, and submitting quarterly performance reports to regulators.
Think of it like...
Post-market monitoring is like ongoing vehicle safety inspections after a car leaves the factory — just because it passed quality control on day one doesn't mean it's still road-safe a year later.
Related Terms
Model Drift
The gradual degradation of a model's predictive performance over time as the real-world environment changes. Model drift can be caused by data drift, concept drift, or both.
AI Incident
An event where an AI system causes or nearly causes harm, produces unintended outputs, or fails to perform as expected in ways that affect individuals, organizations, or the public. AI incidents require documented response, root cause analysis, and may trigger regulatory reporting obligations.