With the online proliferation of hate speech, organizations and governments are trying to tackle this issue, without upsetting the `freedom of speech'. In this paper, we try to understand the temporal effects of allowing hate speech on a platform (Gab) as a norm (protected as freedom of speech). We observe that the amount of hate speech is steadily increasing, with new users being exposed to hate speech at a faster rate. We also observe that the language used by the Gab users are aligning more with the hateful users with time. We believe that the hate community is evolving a new language culture of their own in this unmoderated environment and the rest of the (benign) population is slowly getting adapted to this new language culture. Our work provides empirical observations to the HCI questions regarding the freedom of hate speech.
View on arXiv