Internal Facebook documents show the company’s censorship algorithms give users broad latitude to those who write hate speech, especially if those attacks are targeted at certain “subsets” of “protected categories,” ProPublica reports. For example, protected categories are based on race, sex, gender identity, religious affiliation, national origin, ethnicity, sexual orientation, and serious disability or disease. While “white men” are protected under these rules, “female drivers” or “black children” are not because one of their descriptors is not protected. Facebook said it aims to apply consistent standards of protection worldwide.
