Deep diving into the topic of online hate speech with SELMA at the Safer Internet Forum 2019

On Thursday, 21 November 2019, the Safer Internet Forum (SIF) took place in Brussels, Belgium. With a theme of "From online violence to digital respect", it also celebrated 20 years of safer/better internet funding by the European Commission. Below, read the summary of a deep dive session on online hate and the SELMA project, led by Ken Corish, Online Safety Director at South West Grid for Learning (SWGfL), Stefanie Fächner, Media Education Consultant at the German Awareness Centre and Niels-Christian Bilenberg, Helpline Coordinator at Cyberhus.


In this session, Ken Corish, Stefanie Fächner and Niels-Christian Bilenberg shared information about the SELMA project and its pedagogy. Online hate speech is a difficult issue to discuss with young people; it is complicated to understand and to distinguish from other forms of harmful content.

Ken Corish opened the session by presenting SELMA, which, according to him, is not just another resource, because of its highly active and customisable components. The SELMA project has a Social and Emotional Learning (SEL) background, which is part of the SELMA acronym (Social and Emotional Learning for Mutual Awareness). The motto of SELMA is "Hacking hate". The hacking flow, according to Ken Corish, is to identify, acknowledge, explore, understand, create, apply, disrupt and change – SELMA aims to teach this to young people as they need strategies to stand up to hate speech.

Niels-Christian Bilenberg gave participants a definition of what online hate speech consists of. Online hate speech is defined as "any kind of statement expressed and spread by an individual or group of people through any form of digital communication targeting an individual or group of people based on a core characteristic of them with the intention to spread hate, harass, threaten and provoke direct or indirect violence" – in short, "any online content targeting someone's core characteristics with the purpose of spreading hate, threatening or provoking violence against them". These core characteristics depend on a person's biological gender, gender identity, race, colour, ethnicity, religion, sexual orientation, or disability.

Niels-Christian Bilenberg then highlighted how the sheer amount of content uploaded to social media platforms every day makes it simply impossible to imagine all of it being checked by moderators. Indeed, every day, 95 million Instagram posts, 350 million Facebook comments, 500 million tweets, 3.5 billion snaps, 60 billion WhatsApp/Messenger messages, and 600,000 hours of YouTube videos are uploaded. Therefore, Niels-Christian Bilenberg introduced the SELMA algorithm for checking hate speech, as shown below.

Representation of the SELMA algorithm of hate

Following this presentation, participants were divided into groups, with each group becoming an algorithm to try to determine whether a specific piece of content is hate speech or not. This activity saw lively group discussions around seven images of online communication for groups to decide whether something is hate speech. There were disagreement within and between groups regarding what a protected characteristic should be and what qualifies as intent to incite hate. Ken Corish and Niels-Christian Bilenberg underlined how difficult it is to give an answer to certain questions and that it is not necessarily about being wrong or right, but about discussion.

They then went on to explain the importance of hacking/disrupting hate. Ken Corish gave important points of intervention, like identifying the context, what medium to use to defuse the hate, and so on. A short exercise followed, with Ken Corish advising participants to tackle online hate speech using

Ken Corish then summarised the session by emphasising the collaborative aspect of hacking, giving the example of Greta Thunberg and the Parkland shootings. He also repeated the importance of SEL in this project and invited people to explore

For more information about the Safer Internet Forum 2019 "From online violence to digital respect", you can read the full report on and visit

Related news