Researchers have issued a stark warning that seemingly harmless AI-generated clips of fruits and vegetables committing graphic acts of violence are spreading extremist agendas on TikTok and pose a serious 'radicalising potential' for young people. The findings, published by the Global Network on Extremism and Technology (GNET), highlight a disturbing trend where these videos circumvent the platform's moderation policies that prohibit extreme violent material.
The rise of violent AI 'slop'
Unlike the viral soap-opera style videos earlier this year featuring strawberries and bananas in paternity drama, these new clips are far more sinister. Masked by a cute aesthetic, some videos carry messages from extremist organisations such as the Islamic State and extreme right groups, and feature what experts describe as 'nihilistic violence extremism'. The content is part of a broader rise of AI-generated clips of anthropomorphic foods that play out melodramatic romance, betrayal, and scandal plots, but with a violent twist.
GNET experts stated: 'Violence-based AI slops reinforce a fascination with violence for violence's sake, thereby allowing the so-called online 'ode to violence' to progressively gain ground among young people within the digital space. More concerning, however, is the escalation from absurdist AI-generated drama into explicitly graphic content.'
Normalisation of cruelty
The radicalising potential of these videos is not just in the ideology, GNET warned, but in the normalisation of cruelty. 'The transition from harmless surreal humour to graphic AI gore reflects a broader dynamic of online radicalisation: the gradual erosion of emotional boundaries through constant exposure to increasingly violent and dehumanising imagery,' they added.
In one disturbing example seen by Metro, a potato holds a knife as a carrot is tied to a tree, then slits the carrot open as blood pours out. Another features a 'Carrot Khalifah' telling his army 'the Carrot State will rise again', which experts said evokes an infamous slogan used by IS. Two IS posts included hashtags often used by the terrorist group's supporters to disseminate propaganda.
Links to extremist groups
The report also highlighted an account that shares violent fruit and vegetable AI content and references neo-Nazi terrorist organisation Atomwaffen Division, as well as the nihilistic violence subculture No Lives Matter. Comments on the content include statements of extreme right-wing ideology, researchers said, with the message 'potato supremacy' referring to white supremacy.
The team raised fears that users may 'stumble' on the terror material and approach membership in the online ecosystems. The authors called on TikTok to develop 'countermeasures' to effectively limit such AI-generated content. Metro has contacted TikTok for comment.
Implications for child safety
This development underscores the challenges social media platforms face in moderating AI-generated content, especially when it is designed to bypass filters. Parents and educators are urged to be vigilant about the content children consume online, as the line between harmless entertainment and extremist propaganda becomes increasingly blurred.
The GNET report serves as a critical reminder that AI technology can be exploited for malicious purposes, and that proactive measures are needed to protect vulnerable audiences from harmful content that masquerades as innocent fun.
