AI-Generated Far-Right Persona Monetising Adult Content on Instagram
What We Found
Dor's Research Team identified an Instagram account presenting as a young working-class Australian woman circulating short-form video content combining antisemitic tropes, anti-immigration messaging, and right-wing populist commentary.
Analysis of 30 posts published between January and March 2026 strongly indicates the persona is AI-generated. No verifiable real-world identity, pre-existing social media history, or third-party corroboration exists. Visual consistency across varied settings and a high, formulaic production pace are consistent with generative AI tools rather than individual content creation.
The account is operationally sophisticated and commercially oriented. It functions as a radicalisation funnel, directing audiences from mainstream political grievances toward severe antisemitic content, before converting that audience toward a subscription-based adult content platform.
The account posted (January 12, 2026) a video of the persona sitting on a reclining chair outside while smoking a cigarette. In-image captions state, "Forced to pay half my salary to [Israeli flag emoji], pdf files [i.e. paedophiles] and fat bitches who don't work and have 8 kids.”
The Three-Stage Formula
The account's structure follows a three-stage formula: an AI-generated identity designed to appear credible and relatable, antisemitic and extremist content deployed as an engagement driver, and commercial conversion of the resulting audience.
Entry-level content covers housing affordability and dissatisfaction with the Australian government - mainstream grievances with broad appeal and no immediate extremist signal. This content functions as the top of the funnel, attracting audiences who have no indication the same account also publishes blood libel conspiracy theories.
From there, content escalates through anti-immigrant messaging to severe antisemitic material, including the juice box emoji as a phonetic slur for Jewish people, Epstein-linked conspiracy theories framing Jewish people as predators of children, and Australian taxation framed as involuntary funding of Israel.
The account also employs coded language for moderation evasion, including the phrase "the T is silent" in the context of a rabbit hole - removing the T from "rabbit" produces "rabbi" - framing conspiracy content as a gateway to discovering Jewish influence.
The account's commercial infrastructure directs audiences from Instagram Stories Highlights to an external link aggregator, and from there to a subscription-based adult content platform. The connection between the antisemitic content and the adult content monetisation is direct: the same audience built through extremist engagement is the audience being sold to.
What This Means
Generative AI has reduced the barrier to creating a convincing synthetic influence operation to near zero. No real person or authentic identity is required, and no individual bears ongoing risk. The operation’s design - a three-stage radicalisation funnel, layered moderation evasion using emoji substitution and ironic disclaimers, and deliberate localisation to Australian political anxieties - reflects a level of platform literacy and strategic intent well beyond opportunistic posting.
What makes this account concerning is not any single post but the architecture behind it. Antisemitism is not the goal - it is the tool. The operator uses antisemitic and far-right content to rapidly build an engaged, ideologically primed audience, then redirects that audience toward a commercial product. This is a calculated business model in which hatred is instrumentalised.
This reflects a broader trend: malign actors increasingly exploit social and political tensions - economic anxiety, demographic change, institutional distrust - as entry points for audience growth. Antisemitism is particularly effective, offering simple explanations for complex grievances, driving emotional engagement, and positioning the creator as a suppressed truth-teller. Each dynamic directly translates into reach and follower growth.
The commercial layer compounds the harm. When audiences built on antisemitic engagement are monetised, hatred becomes self-sustaining. Platform bans do not fully disrupt this model: external infrastructure - landing pages, subscription platforms, payment processors - continues to generate revenue independently.
AI generation makes the model significantly more dangerous. A synthetic persona carries no reputational or legal risk, can be replicated across platforms at low cost, iterated rapidly, and discarded when detected. In this dataset, severe antisemitic content - including blood libel tropes, systematic slurs, and conspiracy narratives - was produced and distributed at scale with no human identity attached. The combination of severity, scale, and anonymity represents a substantive escalation in online hate production, one that current enforcement mechanisms are poorly equipped to address.