To be honest, if someone thought that public things on the internet are not getting scraped, I am not sure what to tell them… Search engines have been doing it since the beginning of search engines, it is no wonder that the same would be done to train AI.