Corp.csv, LLC.csv, Name.History.csv, and Officer.csv all claim to be UTF-8, but contain invalid characters. (They're all people's names, and I guarantee you that well over 90% of those people are black or Latino. So.) They can be found via grep -axv '.*' filename.csv.