Radicalization 2025: The New Face of Extremism in Europe
by Chiara Sulmoni, President, START InSight
Abstract In November 2025, Europe observed the tenth anniversary of the Paris attacks, with the Bataclan now standing as the enduring symbol of an era defined by organized terrorism. Back then, the threat was clear and structured: hierarchical networks, trained cells, and operations largely directed by the Islamic State. Today, the picture has changed dramatically. Radicalization processes and various strands of extremism have become far harder to map or predict. They are increasingly fueled by loose, fluid digital subcultures and ideologies drawn from many different sources – often remixed, blended, or reshaped through imitation and viral spread. What emerges is a sprawling, decentralized constellation of violent impulses, symbols, and narratives that drift along the internet’s fringes, steadily pulling in more and more adolescents.
Teenagers and Radicalization: An Uncontainable Phenomenon? ‘Switzerland’s Security 2025’- the Situation Report from the Federal Intelligence Service (FIS) – states that “the number of cases of minors and young adults becoming radicalised online and developing terrorist intentions will continue to increase (1).” For at least six years, the phenomenon of teenagers engaging in various forms of violent extremism has been widespread across Europe, while also affecting other countries such as the United States and Australia. Moreover, several Asian nations—including the Philippines and South Korea—have recorded a rise in right-wing and misogynistic ideologies (2). Youth radicalization thus emerges as a social epidemic that transcends borders, cultures, and ideological lineages. In Michigan and New Jersey, in November 2025, several youth in contact with one another were arrested and charged with planning a terrorist attack in the name of ISIS over the Halloween weekend (3). In the same period, in Canberra (Australia), a seventeen-year-old was detained on suspicion of planning attacks inspired by racist and extremist ideologies; he reportedly wrote in a private chat that these plans provided him a sense of purpose during a period of depression (4). In a February 2025 speech, the head of Australian intelligence highlighted a disturbing set of cases: minors allegedly sharing videos of beheadings in school courtyards, and a twelve-year-old reportedly expressing the intent to blow up a place of worship. Today, the average age at which minors first come onto the radar of Australian intelligence services is 15. He further emphasized how a generation raised entirely online will soon reach the age most vulnerable to radicalization, and how for many of these young people, the digital world has become the primary reference point for identity formation, belonging, and perception of reality (5). In the United Kingdom, similar trends have been observed earlier than elsewhere. Between April 2024 and March 2025, in England and Wales, the 11–15 age group continued to be the most frequently referred for suspected radicalization (6). In 2024, the categories used to classify cases were updated to include entries such as “fascination with extreme violence or mass casualty attacks,” designed to capture situations where no clear underlying ideology can be identified, but an obsessive interest in violent acts emerges. Most cases fall precisely into this gray area, characterized by a general vulnerability to radicalization. The fact that only a limited number of these referrals are subsequently selected for further, structured intervention suggests that they may reflect psychological or personal distress more than the onset of genuine trajectories toward violent extremism. However, the case of Axel Rudakubana -the seventeen-year-old who, in 2024, killed three young girls at a dance school in Southport (UK)- highlighted the limits of risk assessments based solely on the presence, or absence, of ideology. Despite being flagged multiple times for his interest in violence, he was not taken on by the prevention system because no clear ideological motive emerged—a decision that, in hindsight, proved inadequate. In the Czech Republic as well, radicalization of minors occurs primarily online and does not necessarily involve ideological adherence. Police intervene on average ten times a week in response to violence in schools (7). Across the EU in 2024, approximately one third of those arrested for terrorism-related offenses were under 20 years of age. In Belgium, roughly one third of individuals who planned attacks over the past three years were minors (8); in Italy, in July 2025, State Police carried out twenty-two searches targeting adolescents aged 13 to 17 linked to various extremist milieus (9). Between January and November 2025, France recorded 17 minors charged with terrorism-related offenses, two of whom allegedly planned attacks against the Eiffel Tower and Parisian synagogues (10); meanwhile, three young women aged 18–20 were arrested on charges of preparing a jihadist attack in the capital (11). In response to this troubling trend, at the beginning of 2025 the National Anti-Terrorism Prosecutor’s Office considered establishing a section dedicated to minors, in order to study and prevent early radicalization more effectively (12). The list can continue. In Germany, in May 2025, five young people aged 14–21 were arrested for belonging to the far-right terrorist group Letzte Verteidigungswelle (“Last Line of Defense”), which allegedly seeks to overthrow the democratic system by targeting political representatives and facilities housing asylum seekers (13). Membership reportedly requires proof of having committed ideologically motivated crimes, such as assaults on migrants or violent symbolic acts (14). More broadly, right-wing extremism in Germany is increasingly visible. Racist and antisemitic attitudes, swastikas, and Nazi salutes are on the rise among young people, even within schools (15). Symbols and aesthetics associated with the far right are being absorbed into youth culture, contributing to gradual normalization (16). The neo-Nazi milieu reinforces this dynamic by organizing informal events that promote white supremacist music and merchandise, thereby rendering its symbols publicly visible and transforming them into markers of recognition, identity, and belonging (17). Switzerland is grappling with similar dynamics. In spring 2025, a suspected Islamist-inspired knife attack planned by an eighteen-year-old was thwarted (18). In the Canton of Vaud, data from the Unit for the Prevention of Radicalization (UPRAD) indicate that nearly half of the most complex cases involve minors (45%), including children as young as ten (19). Notably, girls are affected to the same extent as boys, contrary to trends elsewhere (20). According to Serge Terribilini, head of UPRAD since January 2026, the most concerning cases involve “13–14-year-olds with a marked propensity for violence, homophobic, antisemitic, or misogynistic discourse, and a high level of engagement on social media and in online video games”(21). In Bern in 2024, the mentoring service supported twelve radicalized individuals aged 11–20 (22). #ReaCT2024 had already mentioned that the then head of Swiss intelligence observed how Switzerland experienced jihadist radicalization among minors to an even greater extent than other European countries (23); more recently, Attorney General Stefan Blättler has described it as a fundamentally social problem (24). While jihadism remains the most lethal form of extremist violence, an investigation by Republik magazine sheds light on an equally alarming, emerging phenomenon, in Switzerland and elsewhere: that of adolescent gangs self-identifying as “pedophile hunters.” By posing as minors online, they lure victims to isolated locations, assault them, and film the attacks (25). Often, the perpetrators are themselves minors -for instance, in Ticino in 2024, a group of 19 youths aged 13-18 was apprehended- (26). These groups may exhibit attitudes, symbols, narratives, and role models associated with the far right, a rhetoric of hate, vigilante justice, and, at times, purely criminal intent. Experts regard this combination as a significant risk to public security and a potential pathway to youth radicalization.
The Intersecting Trajectories of Crime and Extremism The risks outlined above form part of a broader dynamic: in recent years, the entanglement of criminality and extremism has intensified, alongside an increasing exploitation of young people and minors. Of particular concern is the notorious “764,” an international network—classified by Europol among so-called digital cult communities—active on gaming platforms and social media such as Telegram, Discord, and Roblox. The network is fragmented into countless decentralized cells whose members nonetheless share the same forms of deviance and modus operandi: they manipulate young and vulnerable individuals through psychological pressure, threats, and online stalking, coercing them into filming themselves in sexually explicit and humiliating situations. This material then becomes a tool of blackmail: under the threat of its dissemination, victims are progressively forced into ever more extreme behaviors directed at themselves or others, including self-harm and suicide attempts. Children as young as nine have become ensnared in this spiral (27). Conceived by a fifteen-year-old from Texas but now spread globally, 764 is described by U.S. authorities as a “nihilist” group with accelerationist traits. It does not pursue a coherent political ideology but seeks to generate chaos, suffering, and destruction as a form of domination, combining this aim with satanistic and apocalyptic symbolism. Within the network, graphic images and videos are traded as a form of “currency” to gain influence. In this context, violence becomes embedded in dynamics of power and domination, often exercised by adolescents themselves. In its 2025 annual Report, Europol mentions the growing presence of far-right elements within these violent, occultist communities. In Italy, in early 2025, a fifteen-year-old was stopped by the Anti-Terrorism Section of the Bolzano Provincial Police Authority on suspicion of being affiliated with the network, and of intending to kill a homeless person or a person with disabilities in order to disseminate the video online (28). In the United States, a 21-year-old arrested in November 2025 allegedly not only blackmailed teenagers but also contemplated carrying out jihadist attacks and possessed bomb-making manuals (29). The Institute for Strategic Dialogue (ISD) observed Telegram chats with up to 15,000 users (30). As of November 2025, the FBI was investigating more than 350 individuals believed to have connections to these networks (31). On October 30, 2025, the U.S. authorities indicted for terrorism a young man who had been active in the group since his teenage years and who, among other things, published an online guide on how to identify, groom, and blackmail vulnerable minors with mental health issues (32). Because of the complexity of its dynamics and the fusion of abuse, psychological manipulation, and pseudo-ideological references, 764 is now considered an emerging threat at the intersection of online crime and youth extremism, marked by the centrality and glorification of violence (33). Other transnational networks likewise seek to involve minors, pushing them to commit serious crimes, including assaults and killings (34). These criminal organizations have given rise to a phenomenon known as violence-as-a-service, whereby the execution of illegal activities and violent acts is outsourced to teens, with financial compensation, operational instructions, and logistical support provided. “Recruiters” lure “operatives” via social networks, gaming platforms, and encrypted messaging services, exploiting the language and dynamics of digital youth culture and framing violent acts as missions to be completed, challenges to overcome, or levels to be cleared—much like video games. This mechanism, known as gamification, dulls the perception of real-world risks. Consequently, many adolescents become involved in organized crime without fully grasping the concrete and irreversible consequences of their actions. Criminal networks deliberately target minors, who are more easily influenced and whose involvement minimizes the organisation’s exposure. At the same time, many young people show a strong willingness to participate, drawn by the promise of quick money, visibility, and power—a lure that, though illusory, strongly appeals to those seeking identity and immediate gratification. These dynamics are neither marginal nor episodic. Europol data reveal that minors are now present in more than 70% of criminal domains (35); a Telegram channel dedicated to recruiting underage hitmen had, before its shutdown, amassed around 11,000 users (36). Across all these contexts, violence is progressively normalized within everyday digital environments, fostering the emergence of unstructured forms of radicalization. A major contributing factor is exposure to so-called “gore” content—graphic material depicting terrorist acts, torture, killings, war scenes, and abuse, particularly against women. Such content is often accessible even without actively searching for it, thanks to viral dynamics and algorithmic recommendation systems. It can circulate both on dedicated platforms and mainstream services, appearing in feeds or being shared within private groups. To give an idea of the scale of the phenomenon: an analysis conducted by Human Digital across 24 such websites found that between April 2023 and March 2024, average monthly visits increased by 28.5%, from 29.5 million to 37.9 million (37). Research commissioned by the Australian digital security authority (e-security) indicates that 22% of children aged 10–17 have been exposed online to scenes of real-world violence. While further research is needed in order to clarify the long-term effects of repeated exposure, and its potential links to violent behavior, existing studies show that such exposure can have a significant impact on the psychological well-being of children and adolescents. Potential consequences include anxiety, habituation or desensitization, and avoidance behaviors (38). Internalization of such material can also foster a fixation on violence—now recognized as a relevant factor in pathways to youth radicalization (39). Some perpetrators of school attacks or mass shootings were consumers of gore content, and Europol observes that minors’ interest in this type of action—school massacres in particular—is increasing. A central challenge in examining these dynamics lies in recognizing that, while many young people emerge from psychologically, socially, or familially vulnerable environments that increase their susceptibility to manipulation or exploitation, they may also exercise agency and become conscious perpetrators of violent acts.
Understanding the Epidemic of Extreme Violence in the Age of Lone Actors
Well-being and Mental Health: The New Frontier in the Prevention of Radicalization According to Europol, the combination of social isolation, psychological vulnerabilities and intensive use of digital technologies now constitutes one of the most fertile grounds for early radicalization. In this context, mental health is far from a secondary issue: conditions such as anxiety, depression, loneliness, a sense of worthlessness, or traumatic experiences can make some young people more receptive to polarizing narratives that offer simple, immediate answers to deep-seated needs—such as the desire for revenge, power, recognition, or an outlet for personal frustration. As Clare Allely observes in The Psychology of Extreme Violence, “violence often arises from an individual’s attempt to recover a sense of personal value or meaning that has been lost or threatened”(40). The rapid and violent forms of radicalization we face today are therefore not merely a matter of ideology, but function as mechanisms of psychological compensation. Emerging research and statistical data also point to the presence of neurodivergent conditions—such as autism spectrum disorders—in specific radicalization contexts (notably among lone actors and cases of online self-radicalization). These conditions are neither direct nor causal risk factors for extremism. However, in situations of vulnerability, certain associated traits—such as difficulties in interpreting implicit social norms or other people’s emotions, rigid or categorical thinking, and a strong need for clarity and structure—may increase the appeal of groups and virtual environments characterized by clear rules, sharp boundaries, defined identities, simple language, and an immediate sense of belonging. For this reason, there has long been debate over adopting a public-health approach to preventing radicalization -one that integrates security measures with preventive interventions aimed at promoting mental well-being, supporting vulnerable individuals, and strengthening society’s capacity to safeguard its members, particularly young people. In practical terms, this involves building a flexible and resilient network with those already working on the ground: psychologists, social workers, educators, schools, families, associations/organizations that engage with young people in their daily lives. Strengthening local ties and coordination among these actors enables the early identification of distress and timely interventions, providing support and guidance to vulnerable youths before they drift toward extreme pathways. For such a system to function effectively, it must rest on a shared foundation: up-to-date knowledge of the evolving phenomena, a clear understanding of early warning signs and protective factors, continuous professional training, a common analytical vocabulary, and sustained interinstitutional dialogue. Absent these conditions, coordination remains fragmented, information flows are inconsistent, and the overall capacity for timely and effective intervention is significantly weakened. This approach also requires fostering a culture that views youth distress as a collective responsibility, rather than an individual issue to be handled in isolation by fragmented or stand-alone services.
The digital ecosystem as an identity lab Contemporary extremism increasingly manifests as an “autonomous and emancipated” phenomenon, untethered from hierarchical leadership, traditional organizational structures, or centralized propaganda (41). Extremist narratives -understood as frameworks which define the world, identify enemies and victims, and assign individuals a heroic role, a salvific mission, or an exclusive sense of belonging- now circulate horizontally within digital ecosystems, propelled by the active participation of ordinary users. Often informally and without coordination, these users contribute to the production, dissemination, and gradual normalization of ideas rooted in polarizing, implicitly or explicitly anti-democratic and violent worldviews. The strength of these narratives lies in their ability to operate directly at the level of identity, intercepting deep individual needs and offering vulnerable, crisis-ridden or marginalized individuals -or those in transitional life stages- a sense of power and agency that their everyday lives may fail to provide. This dynamic of collective participation and amplification is facilitated both by platform algorithms—which tend to privilege emotionally intense, controversial, and highly engaging content—and by fluid digital subcultures: virtual communities whose members share not only interests, but also aesthetic codes, language, and behavioral norms. These communities are particularly prevalent on platforms frequented by teens and young adults—such as TikTok, Instagram, Discord, Telegram, Reddit, or 4chan—and often emerge around seemingly ordinary themes: fitness, self-help, relationships, lifestyle, or personal development. While not explicitly extremist, through algorithmic recommendation systems, user interaction, and cross-pollination between adjacent environments, these spaces can trigger gradual processes of cognitive and discursive radicalization, without necessarily translating into direct mobilization or violent action. The case of a fourteen-year-old in Singapore illustrates this phenomenon. Within less than a year, he developed self-radicalization across diverse online environments, ranging from jihadist content to far-right material, the incel subculture, misogyny, and antisemitism. Researchers Yasmine Wong and Antara Chakraborthy highlight how online communities that may appear distinct are in fact porous and interconnected; this interconnection stems from convergent narrative themes, emotional framings, and algorithmic mechanisms that facilitate movement between different content domains. These overlapping spaces act like communicating vessels: exposure to one environment increases receptivity to extremist narratives from others, multiplying and diversifying pathways into extremism that often bypass traditional channels of political socialization (Wong and Chakraborthy, 42). Recurring examples of digital communities that may radicalize include those promoting models of dominant masculinity (the “alpha male”), which can evolve toward radical misogyny; trad wives communities, whose narratives idealizing domestic roles may drift toward antifeminist or supremacist positions; and true crime niches, where fascination with criminal cases can slide into the spectacle of violence and the idolization of perpetrators. The fluidity of these subcultures is reinforced by the absence of rigid ideological boundaries: symbols and language cross-contaminate, mix, and can migrate rapidly across platforms.
TikTok and Radicalization as an Emotional Experience Within this digital universe, TikTok represents a paradigmatic case. Political, religious, and identitarian content circulates alongside motivational or survivalist material, often perceived as harmless or even positive. Messages rarely call for violence directly; instead, they evoke themes of strength, self-defense, dignity, and tradition, portraying the outside world as hostile. Videos emphasizing resilience, personal growth, a “winning mindset,” pride, values, or redemption present themselves as genuine inspiration, tapping into real insecurities—including physical ones—to establish an immediate bond. These materials operate primarily at an emotional level rather than stimulating rational reflection. Short, viral formats, polished aesthetics, evocative music —such as nasheeds in Islamic contexts or remixed popular hits in secular ones— and AI-enhanced creative iconography generate instant attraction. Before viewers fully comprehend the content, identification may occur: they feel energized and inspired without questioning the deeper meaning of the messages conveyed. Symbols, emojis, colors, gestures, and postures function as markers of belonging, readily understood by younger audiences but often opaque to adults. In many cases, ideology is conveyed without being named, using allusion, irony, or ambiguity. Violence is often symbolic, enabling content to evade platform moderation. These materials are not primarily designed to recruit ready-to-act perpetrators; rather, they gradually normalize radical narratives, making them familiar, shareable ‘for fun,’ or attractive as expressions of coolness. Serious intent and operational planning rarely originate on mainstream social platforms; instead, they develop and consolidate in private, encrypted spaces. Although the concept of ‘TikTok jihad’ and the phenomenon of accelerated radicalization on social media reflect genuine concerns, digital platforms themselves do not act as autonomous ideological agents. As tools designed to maximize engagement, they lower the barriers to emotionally intense content and amplify dynamics of imitation and virality. Adolescents, in their turn, are not merely passive recipients; through often unconscious interaction, they can shape their own algorithmic bubbles, but may lack the tools to understand these mechanisms. It is precisely this opacity which constitutes the main risk: content framed as positive or motivational draws users in emotionally, gradually introducing worldviews steeped in antagonism and hostility. In this way, TikTok facilitates radicalization via emotional immersion rather than critical engagement.
Symbolic Violence, Emulation, and Lone Actors Digital platforms thus become spaces for alternative socialization, where radicalization arises less from established ideological frameworks and more from online interaction, emulation, and participation in hybrid communities. Prof. Kevin McDonald observes that “ideologies give way to imaginaries; (…); nowadays, radi-calization is less about joining an organization and more a process of immersion in a networked, multi-authored narrative— increasingly taking the form of a conspiracy or a game—shaped by aestheticization, gamification, and performative staging” (43). Almost like a theatrical space, open to everyone. Ideology does not disappear, but transforms into a framework that allows young people to make sense of their own experience. Personal vulnerabilities and psychosocial stresses -such as social isolation, rejection, discrimination, perceived injustice, lack of recognition, or even burnout (44)- can render individuals especially receptive to simplified narratives. Within these contexts, violence takes on a symbolic role. It shifts from an expression of individual frustration to a narrative tool through which identity, status, and social roles are constructed. Perpetrators may become figures with whom others identify and, in some cases, objects of admiration or even veneration. The case of Luigi Mangione, who in 2024 killed a United Healthcare executive in New York, illustrates this process. A small segment of American Generation Z turned him into a pop icon -‘Saint Luigi’-, interpreting his act of violence as a form of alternative justice and as a response to collective frustration and institutional distrust. John Richardson writes that “pinning down Luigi’s motive misses the point. Luigi’s elusiveness is what matters. For a growing number of Americans who seem to be vibrating with existential anxieties, he became a screen onto which they projected their fears and dreams”(45). Such cases show that violence increasingly takes on a meaning-making function once attributed to ideology: it generates shared significance within fragmented yet interconnected communities. Ideology functions primarily as a framework for rationalizing and legitimizing the act, while violence, through its mediated circulation, operates as a collective mechanism for producing identity and meaning. The 2025 arrest of two British youths, each suspected of attempting to emulate Axel Rudakubana’s attack (46), exemplifies how violence can now spread through imitation, with isolated, shocking acts turning into blue-prints for troubled others.
Conclusion Youth radicalization today is not merely ideological; it is symptomatic of a broader crisis permeating the lives of many young people. Contemporary extremism no longer presents in clear, recognizable forms; instead, it manifests as a fragmented constellation of experiences, emotions, and identities, fueled by loneliness, loss of trust in institutions and in the future, and an online world increasingly serving as the primary reference frame. In this context, ideology does not disappear; it loses its role as a structured collective project and instead becomes a framework that provides immediate personal meaning. Radicalization pathways are increasingly individualized: the collective dimension persists, but often emerges later, representing the point of arrival—rather than the point of departure—of trajectories shaped by experiences of marginalization, frustration, or the search for recognition. Within digital ecosystems, violence can function as a symbolic resource for constructing identity, status, and belonging, taking on a value that extends beyond the act itself. Prevention cannot be limited to content control; it must also engage with the conditions that make these narratives appealing and address the deep emotional and identity-based needs they exploit.
Implications for Prevention and the Public Debate
Legislators and Policymakers -Repression and content removal alone are insufficient. Laws focusing solely on punitive measures or the deletion of online content fail to address the emotional and identity-based roots of contemporary radicalization. -Radicalization is complex, often beginning before a structured ideology emerges or a crime is committed. Understanding the process requires examining the relationship with violence—symbolic, aestheticized, or performative—which can be detectable in online behavior or early warning signals. -Multidimensional risk assessment is essential. Criteria should consider emotional vulnerabilities, fascination with violence, engagement with mythologized or vigilante figures, and early signs of imitation, rather than relying solely on explicit ideological adherence. -Policies benefit from integration across sectors. Mental health, education, local communities, platform regulation, and security measures work best when coordinated to support early interventions. -Prevention before criminality is valuable. Early support and guidance can interrupt trajectories toward extremism, reducing the burden on the justice system and limiting potential harm. Practical Reminder: Laws and repression alone are not enough: investing in early prevention, psychosocial support, and integrated policies enhances overall effectiveness.
Prevention Practitioners (educators, social workers, schools, youth associations) -Early stages often precede explicit ideological adherence. Many radicalization pathways begin with emotional or symbolic signals before ideological conviction emerges. -Weak or ambiguous signals may be present: fascination with violence, aestheticization of violent acts, ambiguous irony, or shared symbols within digital subcultures (memes, reels, challenges). -“Motivational” content can carry hidden risks. Narratives framed as positive or empowering may construct hostile, dichotomous worlds (e.g., “victim → revenge,” “injustice → opposition”). -Addressing emotional vulnerabilities and the need for recognition is critical. Supporting affective resilience, positive belonging, and critical thinking before violence becomes a “narrative resource” helps reduce risk. -Distinguishing between youth violence and violent radicalization is important. Pathways may overlap, but causes, dynamics, and solutions differ; not all youth attracted to violence are radicalized. Practical reminder: Early attention to pre-ideological signals can reduce the risk of escalation and prevent the transformation of violence into shared identity.
The media and Public Debate -Avoid moralistic readings. Reducing everything to “monsters,” “deviance,” or “bad apples” oversimplifies and obscures the underlying social and emotional dynamics. -Avoid technological panic. Platforms do not “create terrorists,” but they can amplify pre-existing vulnerabilities; attention should focus on the conditions that make violent narratives appealing. -Shift the central question. Rather than asking only “Where does the hate come from?” it is also important to ask, “Why do these narratives resonate so strongly today?” This involves examining existential voids, the search for meaning, and the role of algorithms. -Narrative responsibility: Reporting or framing violence can mythologize it, encourage imitation, or turn it into a symbolic resource. Practical reminder: Every public account has the potential to reinforce or weaken the symbolic value of violence. Careful choices in words, images, and framing can help prevent imitation and normalization.
Allely, C. S. (2020). The psychology of extreme violence: A case study approach to serial homicide, mass shooting, school shooting and lone-actor terrorism. Routledge
Sulmoni, C., Estremismo giovane, autonomo ed emancipato, in: #ReaCT2024
Wong, Y., & Chakraborthy, A. (2025, 26 settembre). From banal to extreme: When benign online communities become breeding grounds for the far-right. S. Rajaratnam School of International Studies (RSIS). https://rsis.edu.sg/wp-content/uploads/2025/09/CO25197.pdf
McDonald, K. (2024). Au-delà du paradigme des idéologies : penser les reconfigurations des expériences de radicalisation. In R. Bertrand & T. Renard (Eds.), Expériences autour de la radicalisation et sa prévention (pp. 147–162). Toulouse: érès.
Terrorism as a political and social phenomenon that evolves over time, in step with shifting dynamics of competition among individuals, groups, and states. Contemporary terrorism is not a historical exception; it is the outcome of a long accumulation of conflicts, doctrines, and opportunities. Over the past twenty years in particular, we have witnessed an evolution that has repeatedly changed shape without losing its core function: generating fear, polarisation, and socio-political conditioning with a relatively limited investment. After 9/11, jihadist terrorism acquired a globalised and “systemic” dimension, anchored in transnational networks and in a narrative able to link war, identity, ideology, and political pragmatism. The subsequent phase – marked by Iraq and Afghanistan – turned violence into operational experience and symbolic capital: insurgency, terrorism as technique, tactic, and procedure, and communication as a force multiplier ultimately merged. With the rise of the Islamic State between 2014 and 2017, we saw the shift to a hybrid model: territorial organisation and external projection, commando-style teams and lone actors, coordinated attacks and improvised micro-actions. The contraction of the “caliphate” did not mean the phenomenon’s defeat, but rather its reconfiguration: greater fragmentation, more emulation, and stronger individual initiative. Within this framework, the jihadist threat remains particularly significant because it intersects with the conflict dynamics of international relations and competition in the Middle East and Africa, and because it draws on a radical reading of Islam that turns political events into moral pretexts, cloaking them in religious justification. It becomes even more sensitive in Europe when it grafts onto an individual and group search for identity, fuelled by the cultural opposition of a non-marginal segment of second- and third-generation Maghrebi immigrant communities: a terrain where belonging is also constructed “against” the other, and where polarisation functions as a risk multiplier. We are not, in fact, dealing with a monolithic bloc. Today the jihadist galaxy is fragmented: ideologies, operational priorities, and modes of action diverge, which compels us to read contemporary terrorism also as a social phenomenon, not merely as the expression of clandestine organisations. Forms change, but above all the mechanisms through which violence is produced change: faster activation timelines, lower operational thresholds, and greater permeability to emulation and media-triggered mobilisation. From this follows a reflection that, after two decades of adaptations, can no longer be postponed – and one we continue to insist upon: does it still make sense to define terrorism solely as violence aimed at achieving a political result, assessed therefore in terms of intent? Or is it more useful to read it as the effect of applied violence – namely, as a manifestation capable of generating impact regardless of the chain of command? In other words: is the violent act itself terrorism, even when there is no organisation behind it, or when such an organisation cannot be demonstrated? The centre of gravity shifts, then, from structure to event: terrorism in its manifestation, not necessarily in its organisation. Within the jihadist galaxy, terrorism remains an instrument of struggle, resistance, and domination, but it unfolds along a spectrum of violence that ranges from individual action to organised operations; from inspired violence to emulative violence; up to the insurgent terrorism we have come to know in Afghanistan and Iraq and that, in part, we observe in its manifestations in the Gaza Strip, where the Israeli army confronts Hamas (Bertolotti, 2024a). In this perspective, continuity lies not in the form of the attack, but in its function: compressing security and freedom, pushing the state to react, and turning social fracture into operational space.
Trends and dynamics: numbers are declining, but the terrorist threat persists – an analysis of attacks from 2014 to 2025.
From a quantitative standpoint, jihadist terrorist attacks over the past five years confirm a broadly linear trend, with a noticeable decline in the most recent years, bringing the phenomenon back to pre-ISIS/Islamic State levels. Between 2020 and 2025, 99 successful and failed attacks were recorded across the European Union, the United Kingdom, and Switzerland (12 in 2025). The same total – 99 – was observed in the previous period 2014–2018 (12 in 2015). In the wake of the phase of major attacks in Europe carried out in the name of the Islamic State and, subsequently, in plausible connection with galvanising factors triggered by the Taliban’s seizure of power in Afghanistan and by Hamas’s call to action, 221 jihad-related incidents were recorded between 2014 and 2023. Of these, 77 were attributable to the Islamic State. A total of 271 terrorists took part (7 women), 78 of whom were killed during the attacks. Overall, the toll amounts to 457 fatalities and 2,667 wounded (START InSight database). In 2025 alone, 12 jihadist incidents were recorded: a slight decrease compared with the previous year (15 attacks in 2024), consistent with the 12 attacks recorded in 2023, yet clearly below the 18 registered in both 2022 and 2021. This numerical reduction, however, does not translate into lower risk. The share of “emulative” ac-tions – those inspired by attacks carried out in the immediately preceding days – rises sharply. Emulation increases from 17% of all incidents in 2022 to 58% in 2023 (56% in 2021) and reaches 50% in 2025, bringing the phenomenon back to intensity levels comparable to those of earlier years. 2025 also consolidates a structural trend in the evolution of the threat, marked by the predominance of individual, non-organised, and often improvised attacks, which have progressively replaced the structured and coordinated operations typical of the European urban “battlefield” in 2015–2017. This operational model accounted for 92% of incidents in 2025, while also showing a renewed use of vehicle-ramming tactics against civilian targets – mainly pedestrians – in 17% of cases.
The profile of “European” terrorists.
Jihadism, according to the data, continues to appear as a phenomenon with predominantly male participation. Of the 329 perpetrators recorded, 94% are male; women number 10. The picture is also dynamic: while 2020 saw 3 female attackers involved in terrorist actions, no direct female participation emerges in the five-year period 2021–2025. From a demographic perspective, identified terrorists (men and women) for whom public personal data are available show a median age of 26. This value fluctuates over time: 24 in 2016, 30 in 2019, then rising in the latest observed phase to 28.5 in 2023 and, strikingly, 35 in 2025 – thereby indicating an increase in perpetrators’ age. Looking more closely at the 200 individuals for whom sufficiently detailed biographical information is available, the age distribution provides a more nuanced profile: 7% are under 19 (with the weight of minors declining over time), 38% fall within the 19–26 bracket, 41.5% within 27–35, and 13.5% are over 35. Overall, these data suggest a gradual upward shift in average age within the 19–35 segment, with a notable increase in the over-35 cohort, alongside a contraction in the share of minors involved in attacks over the same time span. As to legal status and origins, the prevailing profile of perpetrators falls within the broad category of “immigrants” (first, second, and third generation), both regular and irregular: 93% of individuals who carried out a terrorist act. In the subset analysed through the START InSight database (155 cases out of 237 terrorists), 45% are regular first-generation immigrants; 28% are descendants of immigrants (second or third generation); and irregular immigrants account for 26%. The latter figure, however, is the one that signals the sharpest discontinuity: it stands at 25% in 2020, doubles to 50% in 2021, reaches 67% in 2023, and then settles at 31% in 2025. The trend confirms a growing presence of first-generation perpetrators; also noteworthy is the 6% share of European-origin citizens who converted to Islam, showing a steady – albeit slight – decline compared to previous years’ averages.
The ethno-national dimension of terrorists in Europe.
In Europe, jihadist radicalisation does not affect populations uniformly; it tends to concentrate within specific national and ethnic components. The data confirm a fairly clear proportional relationship between the composition of major migrant groups and the (direct or familial) background of those who commit terrorist acts: the nationality of terrorists, or that of their families of origin, often mirrors the size and historical rootedness of foreign communities within individual European countries. Within this framework, Maghrebi origins predominate. The ethno-national groups most associated with jihadist adherence remain Moroccan (with significant evidence in France, Belgium, Spain, and Italy) and Algerian (in France). Unsurprisingly, the phenomenon appears more pronounced in Belgium and France, where large Moroccan and Algerian-origin communities have, over time, recorded high levels of youth mobilisation toward jihadist milieus and organisations. In France, for example, a substantial share of terrorists involved in recent attacks comes from families of Algerian and Moroccan origin, consistent with the historical presence and numerical weight of these communities in the country (Bertolotti, 2023 and 2024).
Recidivists and terrorists already known to intelligence.
A second, increasingly relevant indicator concerns the significance of recidivism, also as a consequence of releases at the end of recently completed prison sentences. These are individuals previously convicted of terrorism who strike again after leaving prison and, in some cases, manage to carry out violent actions even within detention facilities. The trajectory is clear: recidivists account for 3% of all terrorists who struck in 2018 (1 case), 7% in 2019 (2), 27% in 2020 (6), 25% in 2023 (3), then falling to 8% in 2025 (1). Although clearly declining, the figure confirms the social dangerousness of individuals who – while temporarily neutralised through detention – often do not abandon the intent to act; rather, they postpone it, waiting for a more favourable operational window. Prospectively, this would imply a higher probability of attacks in the coming years, alongside the release of a growing number of inmates convicted of terrorism offences. Further reinforcing this picture, START InSight also highlights the tendency for terrorist acts to be carried out by individuals already known to European law enforcement or intelligence services. In 2020 such cases represent 54% of the total, and 44% in 2021, 37% in 2022 – figures well above the 10% recorded in 2019 and 17% in 2018. In 2023 the figure rises further and stabilises at 75%, effectively confirming the concerns of institutions tasked with countering the threat, although 2025 ended with a more reassuring figure of 23%. Finally, the profile of individuals with prior detention records (including for offences unrelated to terrorism) shows significant continuity across the period considered: 23% in 2021, slightly down from 33% in 2020, but consistent with 23% in 2019 (28% in 2018 and 12% in 2017). Even in light of a markedly lower 2023 figure (8%) and a near-zero level in 2025, the overall evidence continues to support the hypothesis that detention settings remain risk environments where radicalisation and adherence to terrorism may find enabling conditions.
What is the terrorist threat’s real destructive capacity?
To read terrorism realistically, it must be broken down – without conflating different planes – into three levels: strategic, operational, and tactical. The strategic dimension concerns the use of resources to achieve long-term objectives, those that shape the overall configuration of a conflict and influence state behaviour. The tactical dimension, by contrast, relates to the use of force in the single “engagement” to secure an immediate, circumscribed result. Between the two sits the operational level: the hinge that coordinates, sequences, and combines tactical actions in order to generate effects consistent with the strategic objective. This distinction is essential because, in terrorism, what may appear “small” at the tactical level can be decisive operationally and only marginal strategically. Ultimately, this lens brings the employment of people – more than platforms or means – back to the centre of how military and political effects are produced.
Strategic success: marginal and receding.
At the strategic level, the success of terrorist actions – understood as the ability to generate structural, systemic outcomes (disruption of national and/or international air or rail traffic, mobilisation of armed forces, wide-ranging legislative measures) – tends to disappear over time. The historical pattern shows a progressive erosion in the ability to achieve strategic effects, with fluctuations but a clear trajectory: 75% strategic success in 2014; 42% in 2015; 17% in 2016; 28% in 2017; 4% in 2018; 5% in 2019; 12% in 2020; 6% in 2021. From 2022 onwards, attacks no longer manage to achieve strategic success. Past strategic success remains noteworthy when set against the low organisational and financial investment required, especially where actions are individual or low-complexity. In parallel, overall media attention to attacks has declined. At the strategic level, attacks received international media attention in 70% of cases and national media attention in 93%. Commando-style operations and team-raids, when present, attracted full coverage. This “media success” may have had a direct impact on recruiting would-be martyrs or jihadist fighters, peaking during the periods of highest intensity of violence (2016–2017). Yet the media’s amplifier effect on recruitment tends to diminish over time for two converging reasons. First, there has been a progressive predominance of “low-intensity” actions over “high-intensity” ones: the latter have declined, while low- and medium-intensity actions increased significantly from 2017 to 2021, alongside a marked rise in medium-intensity actions in 2023 (75%) and in 2025 (58%), and a renewed uptick in “high-intensity” actions in 2025 (8%). Second, the public’s gradual emotional desensitisation to terrorist violence – especially in relation to low- and “medium-intensity” events – reduces an attack’s capacity to dominate the information space.
The tactical level: concerning, but not central to terrorism’s logic.
If we take as the primary tactical goal the death of the “enemy” – with security forces identified as the target in 28% of cases – this outcome is achieved, on average, in 48% of cases over the period 2004–2025. However, such a broad time horizon significantly increases the margin of error. Focusing on 2014–2025, a deterioration in the capacity to produce the intended effect becomes apparent: low-intensity attacks predominate and failed actions increase, at least until 2022, when tactical success stabilises at 33%, consistent with the 2016 figure. From 2023 onward, by contrast, a reversal in trend is recorded. Over the last six years, the pattern is particularly illustrative. In 2016, tactical success stands at 31%, with 6% failed acts. In 2017, success rises to 40%, while failure reaches 20%. In 2018, success drops to 33% and failed attacks double to 42%. In 2019, success falls further to 25%, then rebounds and stabilises at 33% in 2020–2022. This trajectory can be read as the combined effect of a reduction in terrorists’ operational capability and the increased responsiveness of European security forces. Although in 2023 the figure returns to 50% of actions achieving tactical success – namely, the death of at least one target – in 2024 and 2025 it drops back to earlier levels, with 20–25% tactical successes.
The real outcome: operational success and “functional blockade”.
The decisive point is that even an “unsuccessful” attack, in terms of producing fatalities, can still generate a significant operational result. Terrorist action tends to saturate and constrain resources (Bertolotti, 2023, 2024): it heavily engages armed forces and police, diverting them from routine activity or limiting their capacity to intervene on behalf of the wider community. It can disrupt or overwhelm health services; restrict, slow, divert, or halt urban, air, and maritime mobility; and hinder the normal conduct of daily, commercial, and professional activities, with a direct impact on the communities affected. In doing so, it reduces technological advantage and overall operational potential and compresses resilience. The damage may be direct or indirect and can manifest even in the absence of victims. Above all, it produces a measurable effect: the restriction of citizens’ freedom. From this perspective, terrorism’s success – even when it does not kill – lies in its ability to impose economic and social costs on society and to shape behaviour over time, in relation to security measures and the constraints introduced by political and public-security authorities. This is what we define as “functional blockade”. Despite the progressive reduction in operational capability, “functional blockade” remains one of the most significant outcomes achieved by terrorists, irrespective of tactical success (the killing of at least one target). From 2004 to the present, terrorism has proven effective in achieving “functional blockade” in 80% of cases, peaking at 92% in 2020 and 89% in 2021. This is a notable result, attained with limited resources, confirming a cost–benefit ratio favourable to terrorism. Yet here too a progressive loss of capability is evident: the achievement of “functional blockade” drops to 78% in 2022, to 67% in 2023, and to 58% in 2025.
Bibliography Bertolotti, C. (2024a), Gaza Underground: la guerra sotterranea e urbana tra Israele e Hamas. Storia, strategie, tattiche, guerra cognitiva e intelligenza artificiale, START InSight ed., Lugano. Bertolotti, C. (2024b), Una fotografia del terrorismo jihadista in Europa: evoluzione storica e operativa, in #ReaCT2024, 5° Rapporto sul Terrorismo e il Radicalismo in Europa, START InSight ed., Lugano, ISBN 978-88-322-94-27-9, ISSN 2813-1037 (print), ISSN 2813-1045 (online). Bertolotti, C. (2023), L’evoluzione del terrorismo in Europa: terrorismo di sinistra, destra, anarchico, individuale, e il ruolo degli immigrati nel terrorismo jihadista all’interno dell’Unione Europea (Analisi di correlazione e regressione), in #ReaCT2023, 4° Rapporto sul Terrorismo e il Radicalismo in Europa, START InSight ed., Lugano, ISBN 978-88-322-94-18-7, ISSN 2813-1037 (print), ISSN 2813-1045 (online).
Tik Tok, social media and radicalisation
by Chiara Sulmoni
Above on this page you will find: “Trapped by TikTok’s Algorithm,” an episode of the program ‘Patti Chiari’ (Radiotelevisione svizzera di lingua italiana, RSI), which offers a 360-degree investigation of the digital platform, featuring reports by Nicola Agostinetti and Valerio Scheggia, along with various studio guests.
It includes a contribution by Chiara Sulmoni (START InSight) on extremist content (from 23′ onward).
Platforms, Algorithms, and Emotional Engagement
In recent years, platforms such as TikTok have established themselves as central environments in the construction of youth identity. They are not merely communication channels, but spaces where young people interpret the world, expand their relationships, and give meaning to their experiences. In this context, extremist content has also evolved: no longer limited to explicit propaganda, it now takes the form of narratives embedded within the platform’s cultural codes, designed to appeal to emotions before critical reflection.
Algorithmic logics play a decisive role: recommendation systems prioritize content capable of capturing immediate attention, favoring strong emotions such as anger, fear, outrage, or pride. This mechanism also rewards controversial content, regardless of its nature, contributing to its spread and visibility. On TikTok in particular, continuous and automated consumption—one video after another—reduces intentional choice and accelerates exposure, leaving little room for reflection.
Aesthetics and Codes of Digital Extremism
The form of content is equally crucial. Rapid sequences, evocative imagery (often generated with the support of AI), personal storytelling, Q&A formats, and motivational speeches are all formats familiar to younger audiences. Politics, religion, identity, and entertainment blend into a continuous stream in which the boundaries between ideological and neutral content become increasingly blurred. Techniques such as bait-and-switch—viral content used as a hook—or the use of trending audio to “mask” or make problematic messages more appealing allow radical narratives to be gradually introduced into seemingly harmless content.
Music is a central element. In Islamist contexts, nasheed—a cappella chants from Islamic religious tradition—create a sense of solemnity, community, and shared destiny, sometimes accompanying calls to jihad. In far-right environments, genres tied to youth subcultures—rap, pop, folk, or electronic music—are often used, frequently remixed with memes and viral content. In both cases, music becomes a tool of identity alignment, fostering a sense of belonging to collective realities that transcend geographical boundaries.
Aesthetics are highly curated and play a crucial role as well. Symbols, clothing, gestures, iconography, visual codes, and the lexicon of digital subcultures—such as those associated with the manosphere—transform extremism into not only an ideological phenomenon but also a cultural and aesthetic one. Within these environments, expressions and symbols typical of incel jargon and the broader manosphere circulate, including references to the “red pill,” the “black pill,” or identity categories such as “alpha” and “beta,” which structure simplified narratives of social and gender relations. This aesthetic is not only codified but also highly attractive: it makes content immediately recognizable, visually engaging, and often emotionally seductive, facilitating its dissemination.
The shared language is immediately understandable to those who possess the interpretive keys, but opaque to parents, teachers, and educators. Younger generations actively participate in this evolution, constantly assigning new meanings to symbols and words, and directly contributing to the construction of their own algorithmic experience. Through their interactions—likes, comments, shares, and viewing time—they shape what is progressively displayed on their screens.
In this context, so-called gamification logics also play a significant role. Violence is often softened, symbolic, or embedded within a playful framework. It may be presented as a “challenge” or a “mission,” following typical video game structures in which actions and objectives are organized into levels and rewards. This produces a perceptual shift: violence is no longer experienced as a real event, but as a performance or a test of value, creating distance from its consequences.
Particularly insidious are layered and subliminal messages. Memes, irony, and ambiguity allow for an initially light form of engagement that can gradually normalize polarizing content. Repeated exposure shapes perceptions of the world without the user being fully aware of it. Phenomena such as “Alt-Jihad” or “Islamogram”—inspired by the “Alt-right” and “Terrorgram” ecosystems within far-right circles—represent a telling example: native digital content that blends jihadist narratives, meme culture, video game aesthetics, and anime references, creating a hybrid form of propaganda adapted to the codes of Generation Z.
The Boundary Between Digital and Real
This aesthetic and narrative ecosystem does not remain confined to the symbolic realm. In some cases, it can contribute to shaping perceptions and behaviors in the real world, particularly when it intersects with conditions of isolation and individual vulnerability. Within this framework fits the case of a fifteen-year-old who, in March 2024 in Zurich, stabbed an Orthodox Jew. The boy, described as highly active online—especially on TikTok and Instagram—interacted with environments linked to a digital Islamist subculture, where he consumed and contributed to the circulation of extremist content.
Similar dynamics emerge in different contexts and on a larger scale, confirming the multiplatform nature of exposure pathways. A notable example is that of a teenager in the United Kingdom convicted of terrorism in 2026, who was part of 25 different far-right online chat groups on platforms such as Telegram, Snapchat, TikTok, and Wire. The boy described this process of building his digital identity as a form of escape from reality.
Another example, also from 2024, comes from several Swiss cantons, where threats of attacks or school shootings were reported, often through writings on school walls. According to authorities and school administrations, in most cases these were not genuine intentions but episodes linked to a trend circulating on TikTok—a kind of viral challenge among young people. Despite the absence of a concrete plan, these incidents still required police intervention and precautionary evacuations. The critical issue lies in the gap between action and intention: even actions born as imitation or viral play produce real effects, generate alarm, and contribute to lowering the threshold of tolerance toward the idea of violence, making it more present and “normalized” in the social space.
Such phenomena rarely remain isolated. More broadly, similar dynamics emerge when global events—such as international conflicts or geopolitical crises—are reinterpreted on a personal level, turning into potential “calls to action.” In these cases, the distance between the global dimension and individual experience shrinks, and events occurring elsewhere may be perceived as a direct impulse to act within one’s own context.
Recent analyses indicate that cases of online radicalization among minors are likely to increase; Swiss authorities also report a rise in the phenomenon. This trend must be understood within a broader context: social media are not merely transmission channels, but formative environments where content exposure, algorithmic dynamics, and identity needs intersect.
Countering these dynamics requires a realistic and multidisciplinary approach. Completely eliminating extremist content is impractical: messages are often ambiguous and capable of adapting their form and language to remain effective and visible within constantly evolving digital environments. At the same time, demonizing platforms is counterproductive, as they also represent spaces for creativity and participation. Moderation and deplatforming strategies are necessary, but they show clear limitations and may push users toward less regulated environments.
Ultimately, online radicalization is not the product of a single piece of content, but of a complex and cumulative ecosystem. Understanding it means recognizing the interplay between technology, society, and individual vulnerabilities—and it is precisely within this space that one of the most significant challenges of our time unfolds.
With this new edition, the Observatory’s research journey continues along the path laid down in previous years, while also taking a step forward. If, in the past Reports, we observed and described the evolution of jihadist terrorism, violent extremisms and anti-system manifestations as dynamic and fluid phenomena that were becoming ever less amenable to rigid categorisation, today we must acknowledge that the picture evolved even further. Terrorism can no longer be analysed solely as an organisational fact or as the expression of identifiable clandestine structures; it must increasingly be interpreted as a form of conflict embedded within a complex ecosystem in which propaganda, trauma, technology, emotional mobilisation and social vulnerabilities are deeply intertwined. #ReaCT2025 is rooted in this awareness. The report retains its dual nature as both a scientific publication and a collective work, the product of researchers, scholars, analysts and professionals from different backgrounds, united by the common aim of understanding phenomena that defy simplification and instead demand a cross-cutting, multidisciplinary and up-to-date perspective. To all of them go both my personal gratitude and that of the Observatory: for their rigour, their dedication, and their ability to read the present without yielding either to alarmism or to superficiality.(Claudio Bertolotti, Director, www.osservatorioreact.it)
INDICE/INDEX (in Italian and English)
Claudio Bertolotti (ITA), La parola al Direttore
Claudio Bertolotti (ENG), A word from the Director
Claudio Bertolotti (ITA), #ReaCT2025 – Il Sommario. Terrorismo, radicalizzazione e minacce ibride tra dati e guerra cognitiva.
Claudio Bertolotti (ENG), #ReaCT2025 – The Summary – Terrorism, radicalisation, and hybrid threats between data and cognitive warfare
Claudio Bertolotti (ITA), Terrorismo jihadista in Europa: traiettorie storiche, dinamiche sociali e trasformazioni operative nell’attuale fase di mutamento globale. Evidenze e risultati dell’Osservatorio ReaCT sul radicalismo e il contrasto al terrorismo
Claudio Bertolotti (ENG), Jihadist Terrorism in Europe: historical trajectories, social dynamics, and operational transformations in the current phase of global change. Evidence and findings from the ReaCT Observatory on radicalisation and counter-terrorism.
Andrea Molle (ITA), Il sovranismo in Europa: cornici narrative e andamenti empirici in Italia e Svizzera
Andrea Molle (ENG), A Laboratory of Identity for the European Right
Andrea Molle (ITA), Profilo di rischio degli autori di violenza politica: Stati Uniti vs Unione Europea (UE-27)
Andrea Molle (ENG), Risk Profile of Political Violence Offenders: United States vs European Union (EU-27)
Francesco Bergoglio Errico (ITA), Analisi della documentazione giudiziaria italiana inerente al terrorismo jihadista dal 2011 al 2025
Francesco Bergoglio Errico (ENG), Analysing Italian jihadism through court records from 2011 to 2025
Barbara Lucini (ITA), Conflitti e violenze: la minaccia radicale in un mondo che cambia
Barbara Lucini (ENG), Conflict and violence: the radical threat in a changing world
Marco Lombardi, Emilio Palmieri (ITA), Minacce ibride e guerra cognitiva: verso una nuova architettura dell’intelligence
Marco Lombardi, Emilio Palmieri (ENG), Hybrid threats and cognitive warfare: towards a new architecture of intelligence
Anna Calabresi (ITA), La guerra cognitiva: trauma, manipolazione e radicalizzazione nell’epoca dell’incertezza
Anna Calabresi (ENG), Cognitive warfare, collective trauma, radicalization: systemic risks and integrated responses
Elisabeth Harnes (ITA) L’adescamento di minori e giovani transnazionali verso l’estremismo islamista violento
Elisabeth Harnes (ENG), Grooming of transnational children and youth to violent islamist extremism
Sören Henrich (ITA), Dalla psicologia forense, una breve nota su come comprendere la violenza estremista nel 2025
Sören Henrich (ENG), From forensic psychology: a brief note on how to understand extremist violence in 2025
Matteo Vergani, Andrea Giovannetti (ITA), Come integrare gli interventi “soft” e “hard” per arginare l’estremismo online
Matteo Vergani, Andrea Giovannetti (ENG), How “Soft” and “Hard” interventions can work together to curb online extremism
Chiara Sulmoni (ITA), Radicalizzazione 2025: il nuovo volto dell’estremismo in Europa
Chiara Sulmoni (ENG), Radicalisation 2025: the new face of extremism in Europe.
Ludovico Camposampiero (ITA), Gli attivisti identitari agiscono come influencer della destra radicale e cercano la normalizzazione politica. Intervista di Chiara Sulmoni.
Ludovico Camposampiero (ENG), Identitarian activists act as influencers of the radical right and seek political normalization. An interview with journalist Ludovico Camposampiero by Chiara Sulmoni
Mauro Lubrano (ITA), Fermate le macchine: l’ascesa dell’estremismo antitecnologico
Mauro Lubrano (ENG), Stop the Machines: The rise of anti-Technology extremism
Iuliia Iashchenko, Andrea Carteny (ITA), Echi del conflitto: la guerra russo-ucraina, la radicalizzazione online e il reclutamentoper il sabotaggio nell’UE
Iuliia Iashchenko, Andrea Carteny (ENG), Echoes of conflict: the Russo-Ukrainian war, online radicalization and sabotage recruitment in the EU
Antonio Giustozzi (ITA), Perché’ l’ondata terroristica dello Stato Islamico si è fermata nella seconda metà del 2024
Antonio Giustozzi (ENG), Why the Islamic State’s terrorist wave stopped in the second half of 2024
Alessandra Lanzetti (ITA), Al-Shabaab e la strategia di propaganda. Evoluzione mediatica, narrazioni e implicazioni regionali
Alessandra Lanzetti (ENG), Al-Shabaab and the propaganda strategy. Media evolution, narratives, and regional implications
Lilla Schumicky-Logan, Khalid Koser (ITA), I Cuccioli del Califfato: perché rimpatrio e riabilitazione sono cruciali per la sicurezza dell’Europa
Lilla Schumicky-Logan, Khalid Koser (ENG),The cubs of the caliphate: why repatriation and rehabilitation are critical to Europe’s security
Tarja Mankkinen, Paul Gill (ITA), EUKH on Prevention of Radicalization. Panel tematico su attori solitari e salute mentale – risultati e lezioni apprese
Tarja Mankkinen, Paul Gill (ENG), EUKH on prevention of radicalization: Thematic Panel “Lone Actors and Mental Health Issues” – results and lessons learnt
Luca Tenzi (ITA), Dai Balcani alla Svizzera: la nuova frontiera della guerra ibrida europea
Luca Tenzi (ENG),The Balkans and Switzerland: the emerging frontline of hybrid conflict in Europe
Magdalena El Ghamari (ITA), Bersagli, non minacce? Mappare i vettori di sicurezza nelle diaspore tagike e cecene in Polonia
Magdalena El Ghamari (ENG),Targets, not threats? Mapping security vectors in Tajik and Chechen diasporas in Poland
Luca Guglielminetti (ITA), La diplomazia delle vittime
Luca Guglielminetti (ENG), The diplomacy of victims
Ahmed Ajil (ITA), Migrazione, sicurezza e religione: ripensare un dibattito controverso
Ahmed Ajil (ENG), Migration, security, and religion: rethinking a contested debate
Cognitive Warfare: How a Violent Event Can Generate an Ecosystem of Hate
by Andrea Molle
A recent article published by the Israel Democracy Institute on the antisemitic attack at Bondi Beach provides a useful point of departure for understanding a dynamic that extends well beyond the individual incident. The analytical value of the case lies not so much in the act of violence itself as in the speed with which the event was immediately absorbed, distorted, and repurposed within a hostile information ecosystem, one already primed to generate disinformation, invert responsibility, and mobilize emotional responses. It is precisely in this transition—from the physical event to its cognitive reprocessing—that the geopolitical dimension of MDMH becomes most visible (MDMH stands for “Misinformation, Disinformation, Malinformation and Hate Speech”).
Within the MDMH paradigm, violence is no longer a terminal point but a trigger. The attack does not close a cycle; it opens one. The physical act produces an emotional spike that is immediately exploited to saturate the information space with alternative narratives—often contradictory, yet united by a common function: preventing the stabilization of facts. In this sense, disinformation does not primarily aim to persuade, but to disorient. Truth is not replaced by a coherent counter-truth; it is dissolved into a plurality of competing versions that make a shared reconstruction of reality impossible.
Within the MDMH paradigm, violence is no longer a terminal point but a trigger. The attack does not close a cycle; it opens one
This mechanism lies at the core of
contemporary cognitive warfare. The objective is not adherence to a specific
narrative, but the erosion of trust in the very process by which reality is
established. When every event is instantly reframed as an “operation,” a
“provocation,” or a “false flag,” the outcome is not informational pluralism
but cognitive paralysis. Citizens no longer know what to believe and, in this
state of permanent uncertainty, become more vulnerable to emotional,
identity-based, and Manichean frames.
A decisive element highlighted by the Bondi
case is that this dynamic is not spontaneous. It is not the product of
disordered reactions by isolated users, but of structured information
ecosystems, characterized by cross-platform synchronization, repeated reuse of
decontextualized content, algorithmic amplification, and continuous migration
of the same narratives across social media, private messaging channels,
short-form video platforms, and generative AI tools. In such an environment,
the distinction between information, opinion, and propaganda becomes
functionally irrelevant.
From a geopolitical perspective, MDMH represents a form of low-intensity but high-persistence conflict, in which state and non-state actors can operate at low cost, with plausible deniability and significant cumulative impact. Antisemitism, in this context, is not merely a historical prejudice resurfacing periodically, but a particularly effective cognitive vector: it is emotionally charged, easily recognizable, cross-cultural, and immediately deployable to explain complex events through simple schemas of blame and malign intent.
In such an environment, the distinction between information, opinion, and propaganda becomes functionally irrelevant
One crucial aspect of the analyzed case
concerns the role of artificial intelligence. When automated systems
participate in the real-time synthesis of events—summarizing, answering,
suggesting interpretations—they become cognitive actors in their own right.
They are not neutral tools: they operate within an already polarized
environment and can amplify its distortions. Errors, omissions, or improper
associations do not merely produce misinformation; they contribute to shifting
moral responsibility, rewriting context, and legitimizing hostile narratives
under the appearance of technical neutrality.
This introduces a significant rupture with the past. In classical propaganda, intentionality was identifiable and attributable. In MDMH, manipulation is often emergent, distributed across human actors, algorithmic incentives, and automated processes. Responsibility fragments, while political effect remains. It is a form of power that does not impose an official line, but instead configures the environment in which all lines become possible—and therefore equivalent.
When automated systems (e.g. AI) participate in the real-time synthesis of events—summarizing, answering, suggesting interpretations—they become cognitive actors in their own right
From this follows a fundamental strategic
implication: regulatory neutrality is no longer sustainable. Treating the
information ecosystem as a purely private space, self-regulated by commercial
platforms, amounts to relinquishing any form of cognitive security. Within the
MDMH framework, algorithmic design, content moderation, recommendation systems,
and generative AI are not neutral technical choices but decisions with
geopolitical consequences. They shape which events emerge, which emotions are
activated, and which groups are perceived as threats.
The Bondi case illustrates how a single
local episode can be immediately integrated into a global narrative of hate,
and how this integration occurs faster than factual verification. This reverses
the traditional analytical sequence of “event → interpretation → reaction.” In
MDMH, interpretation precedes the event because the frame is already in place.
The event merely serves to fill it.
From this perspective, speaking of security without incorporating the cognitive dimension is analytically insufficient. The protection of communities, the prevention of radicalization, and the stability of pluralistic societies increasingly depend on the ability to defend the information space as critical infrastructure. This is not a matter of censorship, but of recognizing that freedom of expression presupposes an environment in which facts are at least able to attempt to emerge before being overwhelmed by manipulation.
From this perspective, speaking of security without incorporating the cognitive dimension is analytically insufficient
MDMH therefore forces us to reconsider the
relationship between violence, information, and power. Attacks are no longer
merely acts of force, but cognitive operations designed to produce indirect,
durable, and difficult-to-attribute political effects. Ignoring this dimension
means continuing to respond to a twenty-first-century conflict with
twentieth-century categories.
Designating Antifa: Legal, Strategic, and Policy Implications
by Andrea Molle in the United States
The Trump administration’s recent decision
to designate Antifa as a terrorist organization raises significant questions
about the use of counterterrorism instruments in the domestic context. Unlike
foreign groups traditionally subject to such designations, Antifa is not a
structured entity with at least a minimum level of centralized leadership,
identifiable membership, or a coherent financial apparatus. It is best
described as a decentralized social movement characterized by self-described antifascist
ideology, localized networks, and diverse tactical repertoires ranging from
peaceful protest to violent confrontation. This structural ambiguity is central
to the challenges and controversies surrounding the designation.
From a legal standpoint, the move enters contested territory. Federal law grants clear authority to designate foreign terrorist organizations under Section 219 of the Immigration and Nationality Act. No equivalent domestic framework exists, even though both the Obama and Biden administrations considered domestic terrorism a top national security priority. The Obama administration’s 2011 National Strategy for Counterterrorism explicitly acknowledged the potential threat of ideologically motivated violence within the United States, though it stopped short of proposing a formal designation regime. Instead, it emphasized community engagement and counter-radicalization initiatives. The Biden administration, in turn, released in 2021 the first National Strategy for Countering Domestic Terrorism, which identified racially or ethnically motivated violent extremism, as well as anti-government and anarchist movements, as pressing challenges. That document directed resources to intelligence sharing, law enforcement coordination, and prevention programs but, crucially, reaffirmed that existing U.S. law provides no mechanism to designate domestic groups as terrorist organizations in the same way foreign entities can be listed under Section 219 of the Immigration and Nationality Act. This continuity underscores the structural gap: administrations of both parties have recognized the salience of domestic extremist violence yet have not sought to create a domestic designation framework, largely due to constitutional and political constraints.
Antifa is best described as a decentralized social movement characterized by self-described antifascist ideology, localized networks, and diverse tactical repertoires ranging from peaceful protest to violent confrontation. This structural ambiguity is central to the challenges and controversies surrounding the designation.
Applying the “terrorist organization” label to an informal domestic movement would also probably rely on reinterpretations of existing statutes, such as those covering material support or conspiracy. The First Amendment sharply limits the scope of government action in this area: expression of political views, even radical or offensive ones, is protected speech. To withstand judicial scrutiny, prosecutions would have to demonstrate concrete involvement in acts of violence or provision of material assistance to illegal activities. This high evidentiary bar limits the practical enforceability of the designation.
Strategically, the designation offers
nonetheless certain advantages. It signals deterrence, both to participants and
to those contemplating financial or logistical support. It expands the range of
investigative tools available to law enforcement, including enhanced
surveillance authority and the ability to pursue financing channels. It also
provides a symbolic victory to policymakers who wish to demonstrate resolve
against political violence.
At the same time, the approach carries several
risks. Because Antifa lacks organizational cohesion, the designation may prove
more symbolic than operational. Efforts to prosecute under terrorism frameworks
could generate constitutional challenges and unfavorable precedents. The broad
application of the terrorist label to a movement that includes lawful protest
activity risks chilling legitimate dissent and expanding state surveillance in
ways that may be difficult to constrain. There is also the strategic cost of
misalignment and prioritizing Antifa could divert resources from addressing other
most statistically significant threats.
The policy implications extend to digital environments. Online expressions of sympathy or identification with Antifa could, depending on prosecutorial discretion, be construed as “material support.” Even if courts ultimately narrow the definition, the perception of risk could deter individuals from lawful speech and association, producing a chilling effect inconsistent with democratic norms. This dynamic underscores the tension between counterterrorism objectives and civil liberties protections when tools designed for foreign threats are applied domestically.
The broad application of the terrorist label to a movement that includes lawful protest activity risks chilling legitimate dissent and expanding state surveillance in ways that may be difficult to constrain.
In sum, designating Antifa as a terrorist
organization exemplifies the challenges of adapting counterterrorism frameworks
to movements that are diffuse, networked, and embedded within democratic
societies. The policy yields symbolic and deterrent benefits but faces
substantial legal, operational, and normative obstacles. Its long-term impact
will depend on judicial interpretation, enforcement practices, and whether it
contributes to reducing violence or instead exacerbates polarization and erodes
constitutional protections.
Risk Profile of Political Violence Offenders. United States vs. the European Union (EU‑27) in 2024 – 2025 (first quarter)
by Andrea Molle in the United States
Recent assessments by security agencies and research institutions reveal a growing convergence in the demographic and geographic profiles of individuals involved in political violence across Western democracies. While the specific ideologies vary—racial and ethnic grievance, as well as anti-government conspiracy theories, are more prevalent in the United States, whereas jihadist and separatist movements continue to dominate in Europe—the underlying offender profile is strikingly similar on both sides of the Atlantic. Typically, the individual is a young man, often in his late teens or twenties, who becomes radicalized online and is drawn to act in areas where media exposure is high or where local grievances create fertile ground for mobilization.
The table that follows draws from the most
up-to-date statistics provided by the FBI, the Department of Homeland Security,
Europol’s 2025 TE‑SAT report, the START-PIRUS dataset, and spatial analyses
based on ACLED event tracking. Together, these sources offer a detailed
snapshot of the current threat landscape.
Variable
United States
European Union
Gender
≈ 85 % male
(PIRUS, 1970–2021)
≈ 91 % male
(TE‑SAT 2025 juveniles)
Age
≈ 68 % aged 18‑34 (PIRUS)
> 60 % under 35; 29 % minors (TE‑SAT 2025)
Race / Ethnicity
REMVE white
supremacist actors = 52 % of FBI DVE disruptions FY 2024
Ethno‑nationalist
& separatist actors = 38 % of 2024 attacks
Religion
< 7 % jihadist‑inspired in U.S. plots (HTA 2025); rise in
Christian‑identity fringe
Jihadist ideology behind 24 of 58 attacks; political extremism
scene “post‑religious” (TE‑SAT)
Political affiliation
Anti‑gov/sovereign
+ partisan actors = 49 % of incidents since 2016 (CSIS)
72 % of incidents in metros > 250 k; secondary spike in
low‑density militia counties (arXiv 2025)
Major capitals plus separatist peripheries (Corsica, Basque
Country)
Table 1: Comparative Risk Markers
A clear gender disparity characterizes
individuals involved in acts of political violence across both the United
States and the European Union. In the U.S., data from the START-PIRUS dataset covering
1990 to 2021 shows that approximately 85% of offenders are male. The trend is
even more pronounced in Europe, where Europol’s 2025 TE-SAT report indicates
that 91% of juvenile terrorism suspects arrested in 2024 were male. This
overwhelming male dominance remains one of the most consistent features across
all ideological backgrounds.
Age is another strong indicator. In both
regions, the late teens to early thirties represent the most common age range
for radicalization and mobilization. In the U.S., nearly 70% of offenders fall
between 18 and 34 years old. Similarly, Europol reports that 29% of all
terrorism-related arrests in the EU in 2024 involved minors or very young
adults between the ages of 12 and 20. These figures highlight the growing
vulnerability of younger populations, especially in digital and social media
environments.
Racial and ethnic identity also play a
major role in shaping offender profiles. In the United States, the FBI
classified 52% of domestic violent extremist (DVE) disruptions in fiscal year
2024 as racially or ethnically motivated, with the majority linked to white
supremacist ideologies. In the EU, the picture is more mixed: 41% of completed
attacks in 2024 were attributed to jihadist actors, while 38% were carried out
by ethno-nationalist or separatist groups, particularly in regions with ongoing
autonomy conflicts.
Religious ideology, though no longer
dominant in the U.S., remains a key driver of lethal attacks in Europe.
Jihadist-inspired plots now account for fewer than 7% of DVE cases in the U.S.,
reflecting a broader shift toward secular or hybrid motivations. In contrast,
such plots were responsible for 41% of all fatal terrorist incidents in the EU
in 2024, making religion a more consequential factor on the European front.
Finally, political affiliation has emerged as a defining element of recent violent extremism. In the U.S., data from the Center for Strategic and International Studies (CSIS) shows that 49% of incidents since 2016 involved anti-government, sovereign citizen, or highly polarized partisan actors. Meanwhile, in Europe, the most rapidly growing segment of terrorism-related arrests comes from accelerationist and neo-Nazi microcells—small, decentralized groups often operating transnationally and using encrypted communication platforms to coordinate attacks.
Typically, the individual is a young man, often in his late teens or twenties, who becomes radicalized online and is drawn to act in areas where media exposure is high or where local grievances create fertile ground for mobilization.
In the United States, geographic clusters
of political violence are not evenly distributed. Spatial analysis based on
ACLED event data reveals that California, Texas, Florida, and Georgia lead the
country in the total number of recorded incidents. However, when adjusted for
population size, the Pacific Northwest—particularly Oregon and Washington—ranks
highest in per capita terms. Beyond major metropolitan areas, a a secondary
cluster of concern emerges in rural counties with active militia networks, including
parts of northern Idaho and eastern Oregon. These regions, though less densely
populated, host communities with strong anti-government sentiment and
logistical infrastructure capable of supporting extremist activity.
Across the Atlantic, the European Union shows a similarly uneven pattern. Italy recorded the highest number of terrorist attacks in 2024 with 20 incidents, followed by France with 14. Spain and France also topped the charts in terms of total arrests related to political violence. Meanwhile, low-intensity but persistent acts of separatist violence continue in places like Corsica and the Basque Country, where historical grievances and regional identity continue to fuel localized conflict. These areas remain hotspots for ethno-nationalist activity, even as broader attention shifts to transnational threats.
Religious ideology, though no longer dominant in the U.S., remains a key driver of lethal attacks in Europe
A recent statistical overview based on Europol and START InSight’s data, summarized by Claudio Bertolotti in #ReaCT2024 (annual Report on Terrorism and Radicalisation in Europe) confirms the persistent threat of terrorism within Europe. In 2023, there were 43 attacks executed and 33 more prevented, with over 600 individuals arrested across EU member states. France and Austria topped the arrest figures, reflecting both operational intensity and intelligence prioritization. The data also emphasize the enduring challenge posed by ethno-nationalist and separatist groups—especially in France and Spain—alongside residual jihadist threats. Bertolotti’s analysis reinforces the view that radicalization is increasingly driven by hybrid factors, blending ideology with personal and psychological vulnerabilities, particularly among disaffected youth.
While demographic and geographic indicators
help identify who and where, understanding how individuals radicalize provides
deeper insight into the threat trajectory.
Regardless of ideology, most individuals
who engage in political violence tend to follow a similar radicalization
pathway. Typically, this begins with a personal crisis—such as emotional
distress, social isolation, or financial hardship—that becomes layered with
broader conspiracy theories or identity-based grievances. These narratives
offer a distorted framework through which the individual begins to make sense
of their situation, often blaming institutions, governments, or specific
groups. Over time, the search for meaning or belonging leads them into online
communities where these views are reinforced. Both the FBI and Europol have
flagged social media, gaming platforms, and encrypted messaging apps as key
accelerants in this process, especially among younger users. These digital
spaces provide not only ideological content but also peer validation, making
them fertile ground for recruitment and mobilization.
The likelihood of an individual engaging in
political violence varies based on a combination of demographic, geographic,
and behavioral factors. The highest-risk profile is a male between the ages of
18 and 34, residing in a politically polarized urban area or a region with
active separatist movements. This individual is typically deeply engaged in
extremist content online, often through forums, social media, or encrypted
apps.
A moderate level of risk is associated with individuals living in rural U.S. counties where militia activity is present. This group often includes people with a prior history of minor violence or domestic abuse, suggesting that a background of interpersonal aggression may be a precursor to political violence under political ideological influences.
Regardless of ideology, most individuals who engage in political violence tend to follow a similar radicalization pathway. Typically, this begins with a personal crisis—such as emotional distress, social isolation, or financial hardship—that becomes layered with broader conspiracy theories or identity-based grievances.
At the lowest end of the risk spectrum are
older adults—particularly women over the age of 45—who have no significant
online footprint in extremist spaces. This demographic remains substantially
underrepresented across all known datasets of politically motivated violence.
To effectively reduce the threat of political violence, prevention strategies must target the most vulnerable groups and high-risk environments. One of the most urgent priorities is early intervention aimed at young males between the ages of 13 and 24, who represent the fastest-growing segment among those radicalized online. Outreach programs that engage these individuals before they become deeply embedded in extremist networks can significantly reduce long-term risk.
One of the most urgent priorities is early intervention aimed at young males between the ages of 13 and 24, who represent the fastest-growing segment among those radicalized online.
At the same time, federal and local fusion
centers should align their resource deployment with the geographic clusters
identified by ACLED data and academic spatial models. This means focusing
efforts not only in major urban centers but also in specific counties where
historical or ongoing extremist activity has been recorded.
Importantly, prevention programs should
move beyond rigid ideological classifications. Rather than focusing solely on political
extremists, jihadist, or separatist threats, interventions should be built
around common behavioral patterns—such as personal crises, social isolation,
and online radicalization—that transcend ideological boundaries.
Finally, special attention must be given to
election cycles. Both the Department of Homeland Security and the Center for
Strategic and International Studies have documented consistent spikes in
political violence and threat activity during major elections. Surge planning
and targeted security measures around these periods are essential for
mitigating potential flashpoints.
Estimated Risk to the Public and
Strategic Mitigation
While the political violence landscape is
evolving and increasingly visible, the actual risk of physical harm to the
average citizen remains statistically low in both the United States and the
European Union. Based on aggregated datasets from the FBI, DHS, and Europol,
the annual likelihood of a civilian being killed in a politically motivated
attack is less than 1 in 10 million in most Western countries. For context,
this is comparable to the annual odds of being killed by lightning or a
domestic gas leak. However, these figures mask important nuances. The perceived
threat is far greater in certain high-profile environments—such as government
buildings, political rallies, and religious institutions—where attacks are more
likely to occur, particularly during periods of heightened political tension or
following polarizing events.
Over the last five years, the U.S. has averaged 25–35 documented incidents per year of domestic political violence involving physical harm or lethal intent. In the EU, while the number of successful attacks remains lower, the number of arrests and disrupted plots—over 400 in 2024 alone—indicates significant intent and mobilization potential. The real risk, therefore, is less about mass casualty events and more about the cumulative erosion of public trust, democratic stability, and civic norms.
Prevention programs should move beyond rigid ideological classifications. Rather than focusing solely on political extremists, jihadist, or separatist threats, interventions should be built around common behavioral patterns—such as personal crises, social isolation, and online radicalization—that transcend ideological boundaries.
Three intersecting dynamics heighten risk
exposure for specific groups:
Proximity to symbolic or political institutions (e.g., Capitol buildings, synagogues, embassies)
Demographic visibility (e.g.,
targeted religious or racial minorities)
Participation in high-profile civic activity (e.g., activists, elected officials, journalists)
To mitigate these risks, authorities and
communities must adopt a layered, preventive approach. Key strategies include:
Behavioral Threat Assessment:
Training frontline personnel (teachers, social workers, HR managers) to
recognize early signs of radicalization and intervene before mobilization.
Digital Literacy and Counter-Radicalization: Promoting fact-checking skills, online resilience, and reporting
mechanisms in youth populations—especially males aged 13–24.
Community-Based Partnerships:
Investing in trusted local actors, including faith leaders and
neighborhood organizations, to build relationships and disrupt isolation.
Election Security Surge Planning:
Deploying targeted security resources and disinformation countermeasures
during election cycles, which are now consistently associated with spikes
in threats.
Data-Driven Fusion Centers:
Expanding the capacity of regional intelligence hubs to share geospatial
and behavioral insights in real time between law enforcement and civic
institutions.
Ultimately, while political violence is
unlikely to touch the average citizen directly, its ripple effects can
undermine democratic life if left unchecked. The focus, therefore, should not
only be on physical security but also on rebuilding institutional trust and
strengthening social resilience.
Sources
[1] Europol.
*European Union Terrorism Situation & Trend Report (TE‑SAT 2025)*. Luxembourg:
Publications Office of the EU, 2025.
[2] DHS Office
of Intelligence & Analysis. *Homeland Threat Assessment 2025*.
Washington DC, 2024.
[3] University
of Maryland START. *Profiles of Individual Radicalization in the United
States (PIRUS) Research Brief*, March 2023 update.
[4] Claudio
Bertolotti, ed., #ReaCT2023 – Report on Radicalization and Terrorism
(Rome: START InSight, 2023),
https://www.startinsight.eu/react2023-report-on-radicalization-and-terrorism/
[5] Riley
McCabe. “The Rising Threat of Anti‑Government Domestic Terrorism: What the Data
Tells Us.” CSIS Brief, October 21 2024.
[6] Ravi Varma
Pakalapati & Gary E. Davis. “Spatial and Temporal Analysis of
Political Violence in the United States.” arXiv preprint 2503.14399,
March 2025.
Riots in Los Angeles and the New Front of Irregular Warfare
by Andrea Molle
Irregular warfare (IW) is commonly
understood as a conflict where the struggle is not necessarily over land or
conventional military superiority, but rather over legitimacy, influence, and control
of populations. Traditionally associated with insurgencies, guerrilla tactics,
and non-state actors, irregular warfare has increasingly evolved into a more
complex and hybrid phenomenon, especially in democratic societies. If examined
through this contemporary lens, the tensions unfolding in Los Angeles between
“Angelenos,” local authorities, and the federal government can be seen as a
domestic iteration of irregular warfare.
At the heart of the conflict is a
fundamental contest over legitimacy and sovereignty. Los Angeles, along with
other sanctuary jurisdictions, has actively defied federal immigration
enforcement, refused cooperation with certain Department of Homeland Security
(DHS) directives, and opposed national crime-control initiatives perceived as
unjust or racially biased. These actions reflect not just policy disagreements,
but a deeper ideological struggle over who gets to govern and how. By asserting
local governance norms over federal mandates, Los Angeles effectively
challenges the supremacy of the Federal Government within its own territory—an
act reminiscent of the strategic behavior of irregular actors seeking to
discredit or undermine centralized authority.
Crucially, the methods employed are
asymmetric. Rather than armed resistance, Los Angeles authorities deploy tools
of legal warfare (“lawfare”), bureaucratic resistance, and public messaging.
Strategic lawsuits, municipal non-compliance, prosecutorial discretion, and
ordinances designed to shield undocumented residents are instruments of
resistance, analogous to how irregular forces use terrain, time, and
unconventional means to outmaneuver superior forces. This bureaucratic
insurgency does not seek to overthrow the state, but rather to reshape the
boundaries of federal authority from within.
Yet, the conflict has not remained confined
to the realm of law and rhetoric. In recent days, it has taken a kinetic turn,
as federal agents have clashed physically with protesters, community
organizers, and even municipal law enforcement during raids and enforcement
operations. These confrontations—at times erupting into street riots, mass
detentions, or violent dispersals—mirror the tactical realities of irregular
warfare where control over urban space becomes a proxy for legitimacy. The deployment
of militarized federal units into city neighborhoods without coordination or
consent of local authorities further intensifies the perception of occupation,
leading to spontaneous or organized civilian resistance. This escalation into
physical confrontation blurs the line between law enforcement and political
coercion, a dynamic typical of hybrid conflicts in which the state itself
becomes fragmented and contested.
Equally important is the role of narrative
warfare. Federal authorities characterize Los Angeles as “lawless,” a
city held hostage by crime and chaos, while city officials portray themselves
as defenders of human dignity, civil rights, and moral governance. These
competing narratives are not peripheral—they are central to the conflict, as
both sides vie for public support and political capital. In irregular warfare,
victory often hinges not on battlefield wins but on the ability to sway hearts
and minds. In this respect, the struggle in Los Angeles fits squarely within
the psychological and informational dimensions of IW.
Moreover, this confrontation involves a
complex web of non-traditional actors. Civil society organizations, activist
networks, legal aid groups, and even religious communities have assumed
quasi-political and protective functions, stepping into roles typically
reserved for state institutions. Their coordinated efforts to resist federal
enforcement and provide alternative forms of governance and justice are
hallmarks of irregular conflict, where legitimacy is contested not only through
force but through competing institutions.
In conclusion, while there are no
conventional armies or insurgent militias involved, Los Angeles presents a
modern battlefield of irregular warfare—one where law, identity, narrative, and
at times even physical force are the weapons. As the nature of conflict in
liberal democracies continues to evolve, it becomes increasingly clear that
irregular warfare is no longer confined to distant insurgencies or failed
states. It is now playing out in the contested political geography of cities
like Los Angeles, where the fight for control is not just over policies, but
over the very meaning of sovereignty, legitimacy, and justice in the 21st
century.
Zizians, the rise of the vegan cult: from philosophy to violence
by Andrea Molle.
The recent arrest of Jack LaSota, known
online as “Ziz,” has brought attention to the enigmatic and dangerous
group called the Zizians. Once an obscure internet subculture, it has now been
thrust into the public eye as authorities uncover disturbing details about the
group’s ideology and its links to violent crimes. Under LaSota’s leadership,
the Zizians have evolved from a niche philosophical movement into an organized
and radicalized network, willing to engage in extreme actions to further its
beliefs.
This cult-like organization has been
implicated in violent incidents across the United States, including deadly
confrontations with law enforcement, targeted attacks, and aggressive protests.
Reports suggest that its members adhere to a rigid, almost apocalyptic
worldview, combining radical veganism, anarchism, transhumanism, and a deep
distrust of established institutions. The group’s activities, from online
proselytization to real-world aggression, raise urgent questions about its
origins, recruitment methods, and the broader implications of its growing
influence. As investigations continue, authorities and analysts face the
challenge of dismantling a decentralized movement that thrives on the internet.
The Zizians exemplify how ideological extremism, fueled by online
radicalization, can spill over into real-world violence, posing a unique and
evolving threat to public safety.
The group traces its origins to the San
Francisco Bay Area, where in 2016, Jack LaSota began publishing a blog under
the pseudonym “Ziz.” Initially, LaSota’s writings attracted a niche
audience, particularly within online circles interested in philosophy, artificial
intelligence, and radical social theories. Over time, her ideas evolved into a
more complex and controversial ideology, drawing in followers and forming the
foundation of what is now the Zizian movement. LaSota, a transgender woman,
delved into unconventional theories about human consciousness, proposing that
the brain’s hemispheres could possess distinct values and even separate gender
identities, often existing in internal conflict. She framed this as a
fundamental struggle within individuals, one that could be “resolved”
through personal transformation and ideological commitment. This perspective
resonated with some online communities, but it also fostered a more rigid and
dogmatic worldview among her followers.
LaSota’s discourse expanded into political
and ethical domains, incorporating radical veganism, anarchism, and extreme
opposition to mainstream rationalist communities—particularly those concerned
with artificial intelligence and existential risk. She criticized these groups
for what she perceived as moral cowardice and an unwillingness to take direct
action against perceived threats. This opposition became a defining
characteristic of Zizian ideology, shaping the group’s adversarial stance
against the rationalist movement and its institutions.
The eclectic mix of beliefs that emerged
gave the Zizians a distinct ideological identity. What began as an online
intellectual pursuit morphed into an action-oriented movement, increasingly
militant in nature. The group’s shift from fringe philosophy to violent action
became evident in 2019, marking a turning point in its evolution. That year,
LaSota and several associates were arrested during a protest outside a Northern
California retreat center hosting a rationalist event. What started as an
ideological dispute over artificial intelligence ethics quickly escalated into
direct action, characterized by aggressive tactics such as physical
confrontations, property damage, and attempts to disrupt the event. This
incident signaled a disturbing shift from intellectual critique to militancy,
setting the stage for more extreme actions in subsequent years.
By 2020, the Zizians had attracted
individuals not only ideologically aligned but also willing to engage in
direct, sometimes violent action. In one notable case, a Zizian-affiliated
individual was arrested in Portland, Oregon, after setting fire to a research
facility linked to AI development. The attack, classified as arson, was framed
by the perpetrator as a “preemptive strike” against artificial
intelligence systems they believed posed an existential threat to humanity.
In 2021, a coordinated harassment campaign
targeted key figures in the rationalist and effective altruism communities.
Several prominent researchers received death threats, and at least one
rationalist blogger’s home was vandalized with Zizian slogans. While no direct
physical violence was involved, the campaign demonstrated the group’s
increasing willingness to engage in intimidation tactics.
The escalation continued in 2022, when a
group of Zizians staged a break-in at a biotech laboratory in San Diego,
allegedly to “liberate” animals used in testing. Security footage
showed masked individuals wearing tactical gear, further indicating the group’s
militarization. Though no injuries were reported, the break-in resulted in
extensive property damage, and several members were arrested.
In 2023, violence took a deadlier turn. A
Zizian member was implicated in the attempted murder of a computer scientist in
Boston, a researcher advocating AI safety protocols. The suspect, who had
posted several online manifestos aligning with LaSota’s theories, was
apprehended before the attack could be carried out. However, the incident
reinforced concerns that the Zizians were moving beyond property crimes and
harassment into targeted physical violence.
These incidents paved the way for outright
bloodshed in 2025. In January of that year, U.S. Border Patrol Agent David
Maland was fatally shot during a traffic stop in Vermont. The assailants,
linked to the Zizians, were found with tactical gear and weaponry, underscoring
the group’s operational capabilities and the seriousness of the threat they
posed. Another shocking act of violence occurred in Vallejo, California, where
landlord Curtis Lind was brutally stabbed. Investigations revealed connections
between the suspects and the Zizian network, highlighting the group’s expanding
geographic reach and its growing disregard for human life in pursuit of its
ideological goals.
The pattern of escalation, from online
radicalization to targeted violence, demonstrates the Zizians’ transformation
into a dangerous extremist movement. What started as an obscure philosophical
discourse has now become an organized threat, with real-world consequences that
authorities are struggling to contain.
While the Zizians originated in the United
States, their influence now extends beyond American borders. Their activities
and network have gained footholds in various European countries, sparking alarm
about the group’s global reach and impact. Individuals like German national
Felix Bauckholt, implicated in violent activities associated with the Zizians,
demonstrate the group’s ability to infiltrate and operate across national
borders. Bauckholt’s involvement signals a broader trend of the group’s international
appeal or organization, suggesting a transnational network that facilitates
coordination and ideologically driven violence.
In Europe, authorities are reporting
Zizian-inspired actions in several countries. In the United Kingdom, there have
been instances of radicalization tied to the group’s ideology. France, with its
history of radical movements, has also seen individuals align themselves with
Zizian ideals, heightening concerns about the potential for organized extremist
attacks. Furthermore, countries such as Italy and Spain have become critical
points for recruitment and logistical support, with their porous borders and
diverse political climates making them susceptible to external ideological
movements like the Zizians. These nations’ connections may involve financial
networks, online propaganda campaigns, and logistical support that enable
actions across Europe.
The growing international dimension of
Zizian influence raises several critical concerns. Law enforcement and
intelligence agencies across Europe are increasingly collaborating to track the
group’s activities, share intelligence, and prevent further escalation. The
rise of this transnational extremist network highlights the need for enhanced
cooperation between nations to counter the threat of globalized radical
movements. The Zizians’ ability to inspire or directly coordinate actions
beyond the United States underscores the evolving nature of modern extremism
and the growing complexity of combating transnational threats.
The Zizians exemplify how online
radicalization can give rise to extremist movements in the digital age. Central
to their operations was the digital presence of key figures like LaSota, whose
online platforms became gathering places for like-minded individuals drawn to
the group’s violent ideology. These platforms allowed LaSota and others to
disseminate propaganda, ideologically charged materials, and violent rhetoric,
creating an echo chamber where extremism could flourish without traditional
geographic limitations.
LaSota’s online influence served as a nexus
for isolated individuals who, driven by shared grievances or radical
ideologies, could find solidarity and reinforcement in each other’s beliefs.
This virtual space allowed members of the Zizian network to coordinate actions,
share strategies, and mobilize others toward radicalization. Through encrypted
communications, private chat rooms, and social media platforms, the Zizians
could organize discreetly and spread their messages far beyond what was
previously possible.
The reach of online radicalization cannot
be overstated. For many, exposure to extremist content online acts as a
catalyst for radical thinking, sometimes evolving into violent intentions. The
anonymity and unregulated nature of digital platforms, especially on the dark
web, facilitate the spread of extremist ideologies while bypassing traditional
monitoring and control mechanisms.
This underscores the growing importance of
vigilance in monitoring online spaces to prevent the incubation of such
dangerous movements. The Zizians’ ability to thrive in these virtual spaces
highlights the necessity for law enforcement, social media companies, and
governments to collaborate in identifying and neutralizing online
radicalization efforts. This requires not only enhanced tracking of extremist
content but also a proactive strategy of engagement, where tech platforms are
empowered to detect and remove harmful materials before they gain traction.
Moreover, the Zizians’ use of digital tools
to recruit, radicalize, and plan violent actions illustrates the urgency of
adopting more sophisticated countermeasures in cyberspace. This includes
improving online content moderation, fostering digital literacy programs to
combat extremism, and raising awareness of the dangers of radicalization. The
lessons learned from observing the Zizians and similar groups should drive
global efforts to regulate digital platforms and prevent them from becoming
breeding grounds for violence, helping to stem the tide of online
radicalization before it can manifest in real-world harm.
The emergence of the Zizians, rising from
the shadows of obscure online forums to orchestrating real-world acts of
violence, serves as a chilling reminder of the potential dangers posed by
unchecked extremist ideologies in the digital age. What began as an online
fringe movement quickly evolved into a transnational network capable of
mobilizing individuals across borders, turning virtual radicalization into
tangible, violent actions. This escalation underscores the urgent need to
address the broader issue of how extremist ideologies can thrive in the digital
realm, often going unnoticed until it’s too late.
The Zizians’ ability to recruit,
radicalize, and coordinate violence online highlights significant
vulnerabilities in our current systems of monitoring and regulation. It serves
as a potent warning that the digital space, with its vast anonymity and global
reach, can be harnessed by malicious actors to spread hate and incite violence.
Therefore, it is essential for law enforcement and intelligence agencies to
increase vigilance, while governments and international organizations implement
proactive policies that limit extremist groups’ ability to flourish in these
unregulated spaces.
Policymakers must collaborate with tech companies
to establish robust frameworks for monitoring online content, while also
developing counter-radicalization strategies that are effective and
non-invasive. Social media platforms must take greater responsibility in
identifying and removing extremist materials before they gain traction.
However, these efforts must be balanced with the protection of free speech and
individual rights, which presents a complex challenge requiring nuanced
solutions.
Furthermore, society at large has a vital
role in preventing the radicalization of vulnerable individuals. Educational
initiatives that promote digital literacy, critical thinking, and awareness of
online extremism are crucial in empowering individuals to recognize and resist
harmful ideologies. Community-based efforts to foster inclusion, tolerance, and
dialogue can help counteract the divisive rhetoric that fuels radical movements
like the Zizians.
Ultimately, the rise of the Zizians serves
as a wake-up call, urging us to confront the evolving nature of extremism in
the digital era. The collective responsibility of law enforcement,
policymakers, tech companies, and the public is essential in ensuring that the
digital age does not become a breeding ground for such threats. Only through
continued vigilance, cooperation, and innovation can we hope to combat the
growing menace of online radicalization and protect society from the
devastating consequences of extremist violence.
📌#ReaCT2023 The 4th annual Report on Terrorism and Radicalisation in Europe ⬇📈launches on 23rd May. Don't miss it! 📊📚Numbers, trends, analyses, books, interviews👇 pic.twitter.com/KLIWWlrJXS
🔴📚 OUT SOON! #ReaCT2023 Annual Report on Terrorism and Radicalisation in Europe | Start Insight ⬇ 16 articles by different authors discuss current trends and numbers. Available in Italian and English startinsight.eu/en/out-soon-r…
🔴@cbertolotti1 a FanPage sulle varie ipotesi dell'attacco👉"(...) non si tratterebbe di droni in grado di fare danni significativi, ma piuttosto di una tipologia di equipaggiamento in grado di fare danni limitati con l'obiettivo di portare l'attenzione mediatica sulla questione" twitter.com/cbertolotti1/s…
Per fornire le migliori esperienze, utilizziamo tecnologie come i cookie per memorizzare e/o accedere alle informazioni del dispositivo. Il consenso a queste tecnologie ci permetterà di elaborare dati come il comportamento di navigazione o ID unici su questo sito. Non acconsentire o ritirare il consenso può influire negativamente su alcune caratteristiche e funzioni.
Funzionale
Always active
L'archiviazione tecnica o l'accesso sono strettamente necessari al fine legittimo di consentire l'uso di un servizio specifico esplicitamente richiesto dall'abbonato o dall'utente, o al solo scopo di effettuare la trasmissione di una comunicazione su una rete di comunicazione elettronica.
Preferences
The technical storage or access is necessary for the legitimate purpose of storing preferences that are not requested by the subscriber or user.
Statistiche
The technical storage or access that is used exclusively for statistical purposes.L'archiviazione tecnica o l'accesso che viene utilizzato esclusivamente per scopi statistici anonimi. Senza un mandato di comparizione, una conformità volontaria da parte del vostro Fornitore di Servizi Internet, o ulteriori registrazioni da parte di terzi, le informazioni memorizzate o recuperate per questo scopo da sole non possono di solito essere utilizzate per l'identificazione.
Marketing
L'archiviazione tecnica o l'accesso sono necessari per creare profili di utenti per inviare pubblicità, o per tracciare l'utente su un sito web o su diversi siti web per scopi di marketing simili.
