Online Library

This paper, part of the Legal Perspectives on Tech Series, was commissioned in conjunction with the Congressional Counterterrorism Caucus.

Every day, internet users encounter hateful and dangerous speech online, and some of them choose to respond directly in order to refute or undermine it. We call this counterspeech. Only a few studies have attempted to measure the effectiveness of counterspeech directly, and as far as we know, this is the first review of relevant literature.

We’ve collected and reviewed related articles from a range of fields including political science, sociology, countering violent extremism, and computational social science. These articles do not all use the term “counterspeech,” but they shed light on various features of successful counterspeech, for example, qualities that make speakers/authors more influential in online interactions or the extent to which pro- and anti-social behavior is contagious on the internet.

Recent advances in generative Artificial Intelligence (AI) have inspired interest in using it to counter online hatred. This updated version of the annotated bibliography contains an entirely new section reviewing recent research related to AI-based counterspeech tools.

The literature examining use of online counter narratives to divert potential extremists from a path of violence highlights the importance of messenger credibility, interactive approaches, and the shaping of broader network contexts. However, counter‐narrative efforts have often failed to integrate these insights fully into their digital campaigns. After reviewing several prominent counter‐narrative initiatives in the United Kingdom and United States, this article highlights lessons from offline efforts to counter violent extremism (CVE) that can be used to inform counter‐narrative efforts online. First, CVE efforts benefit from using a networked approach where a range of individuals highly connected in their respective social networks are used to disseminate counter narratives. Second, interactive, interpersonal messaging techniques help to remove obstacles to participation in CVE efforts. Third, counter‐extremism messages and narratives are often better received when they relate to broader community concerns and priorities, rather than focusing exclusively on terrorism or violent extremism. Common challenges faced by both offline and online CVE are outlined, and the important synergies between these efforts examined.

This report represents a distillation of current Social, Behavioural and Economic research findings on violent extremism: What is the cultural basis for violent extremism and radicalization? What motivates individuals and groups to violence, and how is that risk measured? How does our understanding of these findings educate the mitigation of extremism? Can it be prevented or reversed? These and other topics are outlined here to open a forward-looking dialog in the research and policy community that will be crucial to formulating future research direction and for addressing this pressing national concern.

Domestic terrorists pose a significant threat to the U.S., according to the FBI and the Department of Homeland Security. Domestic terrorists often use online platforms—such as social media and gaming—to communicate radical ideas to a wide audience and mobilize likeminded people.

The FBI and DHS have some tools for sharing information with and receiving information from social media and gaming companies on threats related to domestic terrorism. But neither agency has developed strategies and goals related to such information-sharing efforts.

This volume reports on the range of papers presented at the Annual Countering Violent Extremism (CVE) Research Conference 2014 from 7-8 December 2014 in Abu Dhabi, United Arab Emirates. The Conference was organized and hosted by Hedayah (the International Center of Excellence for Countering Violent Extremism), Curtin University, People Against Violent Extremism (PaVE), and the Australian Government, Department of Foreign Affairs and Trade (DFAT). The Conference was also sponsored in part by the European Commission and the United States Department of State. The event was attended by approximately 100 academics, practitioners and policymakers from over 25 countries. The 2014 CVE Research Conference follows from the inaugural CVE Symposium hosted by Curtin University, PaVE, Macquarie University and Hedayah in Perth, Australia in 2013. As the first of its kind in the region, the 2013 Symposium brought together national and international scholars, practitioners, policymakers and former extremists to discuss and debate the current state and future directions for CVE. The intention for the CVE Research Conference is to be an annual event at which the yearly highlights of cutting-edge CVE research and innovation can be presented to academics, researchers, practitioners and policymakers on a global scale.