Overfitting

A modelling error where a machine learning model learns the training data too closely, including noise and outliers, resulting in poor generalisation to new, unseen data.

When an AI memorises the training data instead of learning general patterns. Like a student who memorises test answers but can't solve new problems. The AI aces training data but fails on new data.

Overfitting is a risk management concern because an overfit model performs well in testing but fails in production. Your AI governance framework should require validation practices that detect overfitting before models are deployed.