What Is Concept Drift?
Concept Drift is a change over time in the real-world relationship a model is trying to predict, so that the patterns it learned during training no longer hold.
Concept Drift — a change over time in the real-world relationship a model is trying to predict, so that the patterns it learned during training no longer hold.
Concept drift differs from data drift: data drift is a change in the inputs the model sees, while concept drift is a change in the underlying relationship between inputs and the correct output (for example, fraud tactics evolving so that yesterday's fraud signals no longer indicate fraud). Both degrade model performance after deployment. Detecting and responding to concept drift is a central part of post-market monitoring and the AI life-cycle management required by frameworks such as the EU AI Act and ISO/IEC 42001.
Source: ISO/IEC 22989:2022; NIST AI 100-1
Plain-language explanation
Concept drift differs from data drift: data drift is a change in the inputs the model sees, while concept drift is a change in the underlying relationship between inputs and the correct output (for example, fraud tactics evolving so that yesterday's fraud signals no longer indicate fraud). Both degrade model performance after deployment. Detecting and responding to concept drift is a central part of post-market monitoring and the AI life-cycle management required by frameworks such as the EU AI Act and ISO/IEC 42001.
See where you stand on AI governance
Take the free 7-question maturity assessment and get a personalised action plan.
Free assessment — 3 minutes →