Which consequence arises from applying Min-Max scaling to a feature containing significant outliers?

Min-Max scaling maps data to a fixed range, typically [0, 1]. If extreme outliers exist, the range (max - min) becomes very large, causing all normal data points to be squashed into a very narrow range near zero. This loss of relative variance makes it difficult for gradient-based models to learn effectively. Standardization (Z-score) is generally more robust to outliers than Min-Max scaling, though robust scaling is preferred when outliers are extreme.