Submit incident
Documented

The Anti-Migrant Content Factory Targeting UK Audiences Ran From Sri Lanka on AI and Ad Revenue

January 1, 2025
Curated by Team Raidu · Reviewed by Shiva Ganesh
aiaaic:AIAAIC2131View source ↗
LinkedInX

What happened

In April 2025, a network operating from Sri Lanka was found to be systematically generating AI-powered anti-immigrant and Islamophobic content aimed at British audiences. At the center of it was a Sri Lankan influencer named Geeth Sooriyapura, whose pages flooded UK social media with anti-immigration posts, conspiracy theories targeting political figures including Prime Minister Keir Starmer, Islamophobic content built around "replacement" narratives, and material portraying migrants as an invading force. The posts were engineered for engagement, and the engagement paid.

The content tracked familiar disinformation templates: fearmongering about demographic change, attacks on political leadership, and recurring framing of Muslim communities as an existential threat to British identity. What distinguished this from isolated extremism was scale and systematization. AI generation allowed the network to produce volumes of material no small team could sustain manually, tuned to regional controversies and news cycles to maximize emotional response, clicks, and the ad revenue flowing back to the operation.

The operation was purely profit-driven, not ideological. Sooriyapura was running a content business, and he was open about it. He recruited and instructed students, teaching them to identify divisive topics and monetize the engagement those topics produced through the same platforms distributing the content. The playbook was designed to be replicated. Anti-migrant sentiment was not a cause but a product line.

Meta eventually removed some of the network's pages under its policies on inauthentic behavior. That response addressed the most visible accounts but not the incentive architecture that made the operation viable. The platforms paying out ad revenue while the content ran had no mechanism to flag the operation in real time, and nothing in Meta's enforcement required the network to disgorge revenue already earned or structurally prevented rebuilding under different identities.

The problem this incident reveals is not that harmful content was created but that the systems governing who earns from it leave almost no durable record. When the pages came down, so did any accessible accounting of what they earned, which advertisers funded the runs, and how long the network had operated before outside scrutiny surfaced it. A provable record of what a system did, tied to the specific financial flows that rewarded the behavior, would not have prevented this content from being created. But it would have made the cost of allowing it visible to regulators, advertisers, and platforms before removal became the only accountability tool left.

Reported impact

Affected parties
Not publicly disclosed
Harm type
Not publicly disclosed
Scale
Not publicly disclosed
Financial impact
Not publicly disclosed
Regulatory action
Not publicly disclosed

Classification

Organization
Not publicly disclosed
AI system
Not publicly disclosed
Industry
Not publicly disclosed
Country
Not publicly disclosed
Provider
Not publicly disclosed
Incident type
Not publicly disclosed

Relevant governance controls

Governance control mapping is not available for this record.

  • No controls mappedNot publicly disclosed

Control mapping is analytical. It does not state that any control would have prevented the incident.

Sources and evidence

This record was researched and written by the Index. The event is also catalogued in the following database, which is listed for cross-reference.

AIAAIC Repository
Also catalogued in
The Anti-Migrant Content Factory Targeting UK Audiences Ran From Sri Lanka on AI and Ad Revenue
2025