Elham Asaad Buaras
New reports reveal that artificial intelligence is creating a dangerous new frontier for digital extremism. In both the UK and India, AI-generated images are being weaponized to spread anti-Muslim hatred.
In the UK, a submission to Parliament’s Home Affairs Committee found far-right accounts using manipulated images to reinforce racist stereotypes.
A London School of Economics study of 622 posts showed that content with these visuals was shared 30% more. The research stated that such images, which portray Muslims as a threat, are helping to stoke anti-immigration protests. For example, a fabricated image circulated of a woman in an England football shirt crying as Muslim men jeered. The same networks that spread misinformation during the Southport riots are now using AI to normalize negative stereotypes.
Dr. Beatriz Lopez, who studies algorithmic extremism, explained, “Our research reveals that biases embedded in these systems have serious real-world consequences.” She emphasized that the tools “consistently reproduce harmful stereotypes, associating black, brown, and Muslim people with criminality.” Dr Lopez warned that the normalization of this content is core to the problem, as it now circulates widely without alarm.
A parallel crisis is unfolding in India. A study from the Center for the Study of Organized Hate (CSOH) documented how generative AI is being weaponized to produce anti-Muslim visual hate. The report examined 297 Hindu nationalist accounts and identified 1,326 AI-generated harmful posts. While activity was minimal in 2023, it rose sharply from mid-2024, coinciding with the rapid adoption of AI tools in India. Instagram was the most powerful amplifier, with 462 posts driving 1.8 million interactions. Sexualized depictions of Muslim women drew the highest engagement, at 6.7 million interactions.
In response, The Council on American-Islamic Relations (CAIR) called on social media companies to act against Indian groups using AI and bots to promote bigotry and incite violence. CAIR cited the CSOH report, which found 1,326 AI-generated images and videos that have collectively drawn over 27 million engagements.
This content often sexualizes Muslim women, dehumanizes Muslims, and glorifies violence. CAIR Deputy Executive Director Edward Ahmed Mitchell said, “Social media companies must take concrete action… Fake campaigns like this can and have led to real violence.”
The report issued a stark warning, stating that a small number of accounts have created a significant volume of hate speech in a short time. This proliferation “threatens to further colonize the Indian information sphere” and risks damaging social relations and undermining constitutional secularism.
A critical finding in both the UK and India is the failure of social media moderation. In India, researchers reported 187 posts for violating guidelines on X, Facebook, and Instagram. Only one was removed—a failure rate that highlights the challenge of policing synthetic hate.
