I keep thinking about the arguments around content being used for AI data sets and the arguments around content being archived/offered by sites like Internet Archive. They don’t seem consistent, on either side. Corporations are happy to use data sets scraped from copyrighted content, but they surely don’t want...