For over a decade, digital watchdogs and researchers have meticulously tracked the nefarious activities of hostile foreign states as they deploy divisive rhetoric across social media, aiming to destabilize nations like the United Kingdom. These operations, often tied to geopolitical agendas, have been a primary focus for platforms and policymakers alike. However, a recent, groundbreaking investigation by the Institute for Strategic Dialogue (ISD) in collaboration with The Bureau of Investigative Journalism (TBIJ) has uncovered a disturbing new frontier in this digital battleground. It has identified a sprawling commercial operation, run from Sri Lanka, that is actively spreading anti-migrant and anti-government sentiment throughout the UK. Unlike the state-sponsored campaigns that dominate headlines, this network appears to be driven by a singular, purely profit-oriented motive, demonstrating that the market for chaos and division is not solely state-funded but can be a self-sustaining, entrepreneurial venture.
Through meticulous analysis, ISD and TBIJ uncovered a sophisticated web of more than 100 Facebook pages and groups, all linked to this Sri Lankan–based network. This digital apparatus has amassed a staggering audience of over 1.6 million followers and members, creating a potent vehicle for its content. The posts, often generated using artificial intelligence, are laden with anti-migrant and anti-government messaging, some of which blatantly contravenes Meta’s own policies on hate speech. The most successful single post, a video from a far-right rally, garnered an astonishing 8.3 million views and nearly 23,000 shares, highlighting the network’s immense reach and its ability to inject its divisive material directly into the mainstream UK political discourse. This discovery reveals a dual threat: while politically motivated actors remain a concern, the rise of commercially driven influence operations presents a new challenge. These profit-seeking actors are willing to stoke hatred and polarisation, not for a cause, but simply because it pays.
A deeper dive into the network’s structure and tactics reveals a highly coordinated and deliberately engineered operation, far removed from the spontaneous outpouring of political passion it purports to be. The investigation identified 89 pages and 39 groups, most of which were created or repurposed specifically between April and September 2025, a period of intense political debate around immigration in the UK. Many older pages, whose original purpose was unrelated, were abruptly rebranded to pivot to UK-centric, anti-migrant themes. The operators deployed a suite of tactics to create an illusion of authenticity and organic support, including cross-interaction between pages, coordinated commenting, and the use of Facebook check-ins to feign a local UK presence. They adopted generic patriotic slogans and hashtags and generated AI-produced images of British symbols like Britannia to build a superficial sense of belonging. However, their efforts were betrayed by tell-tale signs of inauthenticity: characteristic linguistic errors, typos, and the use of fabricated UK addresses. This façade allowed a network operating from thousands of miles away to masquerade as a grassroots British movement, successfully infiltrating and amplifying online conversations.
The content propagated by this network is not merely critical of government policy; it is deeply toxic and designed to inflame. The network’s pages frequently used dehumanizing language, describing migrants as “invaders” and echoing the white nationalist “Great Replacement” conspiracy theory. One AI-generated image, purported to depict the future of Britain, showed menacing caricatures of Muslim men on a bus targeting a white woman, while other posts crudely described Islam as “a cancer.” The overwhelming sentiment was anti-Labour and anti-government, with posts calling for Keir Starmer’s resignation, supporting far-right figures, and demanding new elections. The network even disseminated outright falsehoods, including a fabricated video of the Prime Minister’s resignation and a debunked AI-generated clip of him kissing a donor. The impact of this campaign extends beyond the digital sphere, as their content was shared by public figures, including Reform UK politicians and a councillor, giving it a veneer of legitimacy and demonstrating a worrying level of influence over UK political discourse. Politicians from other parties engaged with the content as well, either by liking it or by point out its inauthentic nature, inadvertently driving further engagement.
The investigation traces this entire operation back to a single individual in Sri Lanka who runs an online “academy” dedicated to teaching others how to monetize political content on Meta platforms. This is the crux of the network’s commercial nature. Training videos from the academy’s verified Facebook page explicitly demonstrate how to identify target audiences, leverage AI tools like ChatGPT to find trending topics, and produce viral content, noting that “political niches have the highest potential to go viral quickly.” The operator advises that AI-generated reels achieve virality up to 10 times faster than other content. The primary goal is to exploit Meta’s Content Monetisation tool, earning money from ad revenue based on engagement, views, and plays. This strategy directly contravenes Meta’s policies, which restrict monetization for content that depicts immigration in a polarizing or inflammatory manner and prohibits unlabeled AI-generated content. Furthermore, the network ran paid political ads that were not properly moderated, including calls for a burka ban and AI-generated images of political figures in prison uniforms. This points to a successful, systematic evasion of platform enforcement guidelines, revealing a critical vulnerability in Meta’s oversight.
The implications of this investigation are profound and unsettling. It demonstrates how economically motivated actors can orchestrate a highly effective online political campaign by exploiting the vulnerabilities of major platforms. The widespread availability and ease of use of generative AI tools have lowered the barrier to entry, allowing bad actors to create compelling, high-quality visuals that seamlessly integrate into social media feeds. The failure of platforms to adequately label or remove this AI-generated and hateful content, despite their stated policies, has enabled this network to flourish. While the immediate concern is often focused on state-sponsored influence operations, this case study serves as a stark warning about the shifting information landscape. The market for polarisation and hate is not exclusive to intelligence agencies; it has become a lucrative business for entrepreneurs across the globe. The network’s success, measured in followers and engagement despite its clear inauthenticity, underscores a growing crisis of misinformation and a lack of media literacy among the public. This case reveals that the digital information ecosystem is not just being attacked by states, but is being commercially mined for profit, with the integrity of democratic discourse as the primary casualty.