Help why to apply PCA here

Lets say we have a dataset of 9 dimensional points and I want to apply k-means algorithm.
I was studying an example where they apply PCA before fitting the data into the clustering algorithm. The problem is that I cannot understand why PCA is used in this unsupervised learning setting.

In case we were in a supervised learning setting and we had a labeled dataset I understand that high feauture correlations might lead to bad performance of the resulting predictor. But in this unsupervised learning setting it is not clear to me what’s the use of applying PCA. Could you please assist in this ?