This session will present some of our work on a project called Hate Track, funded by the Irish Research Council and the Irish Human Rights and Equality Commission. In this project, we are trying to develop an online tool that can harvest racist speech from Facebook and Twitter. To do so, we have to develop definitions that can be operationalised in ways understood by computers. This presented a series of challenges as legal, sociological and cultural definitions tend to differ in important respects. With this starting point, this session will go over these definitions, but also present some findings on the notion of virality and the logic of circulation in social media. If hate is difficult to define and circulates everywhere, then how can we stop it? Intense policing of online spaces has had and is likely to have significant consequences for freedom of speech on the one hand and for the creation of trust and cohesion online. Regulating social media platforms has become a crucial political question and a site for power struggle between states and social media corporations. But is total eradication of hate from online platforms even possible? What are the assumptions involved in such efforts? Sidestepping these concenptual conundums this talk will propose a new conceptualisation of digital hate revolving around its socio-political functionalities by asking what is digital hate actually doing? Any regulatory efforts must be aimed towards addressing this question first. The talk will conclude with some of our findings in terms of the performative dimension of online hate and some suggestions regarding relevant policy.