What is a common consequence of a model that generalizes poorly on unseen data?

A model that generalizes poorly on unseen data often struggles to perform accurately when presented with new, previously unseen examples. This is reflected in the model's low accuracy, making it a prominent consequence of poor generalization.

When a model fails to generalize adequately, it typically means that it has not captured the underlying patterns of the data effectively. As a result, when this model is applied to new data, it produces predictions that are often incorrect, leading to lower accuracy. The model may perform well on training data but falters when deployed in real-world scenarios or on validation datasets, where it encounters data it has not been explicitly trained on.

This scenario can be connected to concepts such as overfitting, where the model performs excellently on training data but does not generalize well to new inputs. However, the direct answer reflects the outcome of that poor generalization in terms of accuracy, which is a critical metric in evaluating model performance. Thus, low accuracy serves as a clear and concise indicator of how poorly a model can generalize to unseen data.