Kolmogorov and Smirnov

the Kolmogorov-Smirnov-test quantifies the similarity between two sets of random numbers. could you imagine that two sets of multivariate random numbers (i.e. random vectors) are statistically equivalent in each index, but have different correlation properties?