True or False: A predictor that is highly sensitive to changes in input features is generally more reliable than one that is insensitive, especially in scenarios with limited training data.

False. A highly sensitive predictor often reacts too strongly to small variations in the data, which can cause it to overfit and perform poorly on new examples. When training data is scarce, a model that is too reactive has little evidence to distinguish real signal from noise, so its predictions become unstable. A more robust predictor is one that captures the essential patterns while ignoring irrelevant fluctuations. For example, a linear regression that uses every noisy feature will fit the training set exactly but will predict poorly on new data, whereas a simpler model that ignores minor variations will generalize better.