The training of the Extremely EUnited project was supposed to take place in the European capital, but Covid 19 forced us to review our plans. It was decided that it would be held anyway and take place online, on the Zoom platform, over five days (from 9th to 25th of June). It was quite a challenge considering the circumstances, but the result was pleasantly surprising. We tried to make the training as interactive and dynamic as possible, by combining classical webinars with workshops and direct testimonies. At the first session we had a guest of honour, Daniela Pisoiu, senior researcher at the Austrian Institute of International Affairs and collaborator of RAN (Radicalisation Awareness Network), who gave the introductory lecture. After underlining the need to clearly define the main concepts related to terrorist radicalization such as radicalisation itself, extremism, lone wolf, she illustrated the major findings of research in the field and the possible angles to look at the phenomenon. Usually, the field of international relations and terrorism studies are permeated by the so-called structure-agency debate. One the one hand, deterministic approaches state that structural factors, such as root causes and psychological characteristics, necessarily lead individuals to radicalise, independently from their will or personal political convictions. On the other hand, intentional approaches lead the violent action back to a deliberate choice of the individual, which is supposed to be rational and not bound by any oppressive structure. Yet, there is also a third explanation for individuals’ radicalisation, which relies on the preponderant role of relational factors such as values, beliefs and ideas. At the end of the session, we came at the following conclusions: it doesn’t exist a unique profile of radical, the pathways to violence are multiple and multidimensional, the context as well as the subculture matter and resilience factors are important in a prevention campaign. In the second session, the analysis of two case studies allowed us to observe on a micro-scale the individual processes of radicalisation and detect abstract pathways, which provide essential information on the mechanisms of radicalisation. Furthermore, with the support of RAN videos, we also addressed specific issues such as the process of radicalisation among refugees and the supposed causal link between mental health problems and fanatism, which turned out to be more complex than expected. It was also an innovative experiment, because at the end of the presentations we divided the participants in groups, within which a discussion took place on the topics addressed. We resumed the training the following week with a webinar focusing on the existing academic research on radicalisation, alternative online narratives, good practices and terminology. Differences between Al-Qaeda and ISIS in terms of narrative strategies were also explained in order to give a general overview of the kind of messages that the European youth is exposed to. In the fourth webinar, we focused on the elements making a social media strategy successful, such as the best social channels to reach out youngsters, the language to use on social media platforms and the role of influencers, especially on YouTube. A week later we were ready for the last session, which would be exceptional in some way since we had a special guest: Saliha Ben Ali, a woman of Tunisian origin, whose 14 years-old son lost his life after leaving Belgium and joining terrorist groups in Syria. It was a touching testimony which helped us to remember that radicalization is not simply a vague possibility existing in the European context, but a real, concrete wound that afflicts and divides the social fabric of our societies. During that last training session, we also had the chance to meet an EU project officer, Petar Kovachev, who answered our questions on technical issues concerning the project. The session ended up with a moment of experience sharing between partners and a presentation on the General Data Protection Regulation.