Anomaly Detection is concerned with identifying observations that deviate from some model of normality. In his Structure of Scientific Revolution, Kuhn argues that anomalies are the major driver of scientific progress, since they uncover flaws in current models.
“Hypotheses” or “models” of how the the world works (for example, in the form of machine learning models) can break down when applied outside of their usual domain. When we make decision based on the predictions of such models, we therefore have to ensure that the model is currently working in the domain it is designed for, so that we can expect it to make accurate predictions. This can be achieved by anomaly detection methods.
An example of such a hypothesis would be Newtonian physics, which works remarkably well for low masses and velocities, but is inaccurate outside of these regimes.