Online Library

This report compares two large random samples of Twitter accounts that tweet in English: one taken just before Elon Musk acquired Twitter in October 2022, and one taken three months later, in January 2023. It also examines several related datasets collected during the period following the acquisition, a period in which, the study found, new accounts were created at a record-breaking pace. Some extremist and conspiracy networks created accounts faster than the baseline rate, probably because changes to Twitter’s trust and safety policies had been announced. In the context of these policy announcements, the study examines some reinstated accounts, with mixed results. Despite the loosening of several content policies, accounts that automated the sending of tweets (‘bots’) saw activity drop sharply during the period of the study, with many bot accounts being suspended or deactivated, while others voluntarily curtailed their activity in light of the API changes announced. Deactivated accounts were dominated by sex- related content and apparent financial spam or scams, often coupled with automated tweeting.

It is well established that jihadist groups and their supporters post URLs on online platforms to outlink to items of propaganda stored on other platforms. Industry initiatives – such as Global Internet Forum to Counter Terrorism’s inclusion of URLs in its hash-sharing database, and Tech Against Terrorism’s Terrorist Content Analytics Platform – have sought to counter this practice. These measures, together with new regulatory regimes (e.g. the EU’s Terrorist Content Online Regulation) and the growing use of decentralised services, raise the question whether jihadist groups’ propaganda dissemination strategies are perhaps being forced to evolve. This report considers whether there is evidence of such an evolution, by examining the means that three jihadist groups (Islamic State (IS), Al-Qaeda (AQ) and Al-Shabaab) used to disseminate their propaganda during a two-month period in early 2023.

Over the past decade, two major phenomena have developed in the digital realm. On the one hand, extremism has grown massively on the Internet, with sprawling online ecosystems hosting a wide range of radical subcultures and communities associated with both ‘stochastic terrorism’ and the ‘mainstreaming of extremism’. On the other hand, Artificial Intelligence (AI) has undergone exponential improvement: from ChatGPT to video deepfakes, from autonomous vehicles to face-recognition CCTV systems, an array of AI technologies has abruptly entered our everyday lives. This report examines ‘AI extremism’, the toxic encounter of these two evolutions – each worrying in its own right. Like past technological progress, AI will indeed be – in fact already is – used in various ways to bolster extremist agendas. Identifying the many opportunities for action that come with a range of AI models, and linking them with different types of extremist actors, we offer a clear overview of the numerous facets of AI extremism. Building on the nascent academic and government literature on the issue as well as on our own empirical and theoretical work, we provide new typologies and concepts to help us organize our understanding of AI extremism, systematically chart its instantiations, and highlight thinking points for stakeholders in countering violent extremism.

VOX-Pol Blog post.

VOX-Pol Blog post.

VOX-Pol Blog post.