I don't see anything in that licence that would allow it to be used as a corpus for machine learning

most likely they're relying on fair use, which would apply regardless of where it's hosted