The new partnership will give AI developers access to a dataset 'built with machine learning workflows in mind,' which could help Wikipedia deal with the strain data scraping is putting on its servers ...