Not sure if you're joking or serious but many "modern" natural language processing models like BERT are trained using offline dumps of big collections of text like wikipedia - nothing unusual there and nothing to worry about :)