Alarming rise of AI-generated Islamophobic images on Indian social media: Report

A recent reportpublished by the Centre for the Study of Organised Hate (CSOH) documented the alarming rise of AI-generated images to create an Islamophobic rhetoric in India.

Authored by Nabiya Khan, Aishik Saha and Zenith Khan, the report highlights the role of AI-generated images in shaping anti-Muslim visual hate content in the digital landscape. It states that while there has been an in-depth documentation through studies, reports, and journalistic investigations, the use of AI-generated visuals with the sole purpose of dehumanising Muslims has seldom been studied or analysed.

It studied 1,326 publicly available AI-generated Islamophobic posts between May 2023 and May 2025 from 297 accounts across various social media platforms, with a large engagement.

According to the report, there were four main categories revealed upon analysis: the sexualisation of Muslim women, exclusionary and dehumanising rhetoric, conspiratorial narratives, and the aestheticisation of violence.

The data shows that mass adoption of ChatGPT in the Indian context could result in an explosion of hate-driven content “with grave implications for India’s religious minorities”, not limited to threats, psychological harm, and physical violence.

With an estimated 22 million AI users in India, the report found that even though most of it is used for routine activity, with the growing climate of hate and violence in the country, it is now used as a tool to target religious minorities and caste-oppressed communities.

Also Read

Action against Zee News, Times Now Navbharat for Islamophobic reporting

Nabiya Khan, co-researcher of the report, shares that she saw the inflow of Islamophobic content in the Indian social media discourse as “a symptom and an amplifier of the risks of AI’s rise.” 

The report also examines how such posts are spread and amplified across various platforms by far-right-wing media outlets and networks.

“These tools made old prejudices scalable, faster, cheaper, and harder to trace. Existing laws are insufficient for governing or regulating this content. That’s what pushed us to document it,” Khan was quoted by Maktoob Media.

Whether it be the Assam BJP’s apparent Islamophobic video depicting a dystopian future for Assam or WhatsApp forwards showcasing AI-generated images of the Muslim community in a particular light, the report indicates that AI-generated imagery in India is particularly “worrisome”.

Also Read

Assam BJP shares Islamophobic AI video ahead of 2026 polls, draws flak

While working on the report, Khan shared a key finding that “AI is not creating new hate but automating existing hate.”

It was observed that these images were depicting Muslims as “an inherently violent community” as a tool of propaganda. The report states that the images feed into the wider narrative of Muslims as a “permanent threat”.

One of the key findings showed that, “The category of sexualized depictions of Muslim women received the highest engagement (6.7M interactions), revealing the gendered character of much Islamophobic propaganda, which fuses misogyny with anti-Muslim hate.”

Some images showed snakes wearing skull caps, which further dehumanises the community, framing them as “deceptive, dangerous, and deserving of elimination.”

Additionally, conspiracy theories of “love jihad”, “population jihad” and “rail jihad” were used to frame Muslims as perpetrators and a “threat” to Hindu national security.

The report suggests that such imagery allows Hindu nationalist actors in the political arena to agitate hateful narratives without being fact-checked.

Images using Ghibli-style on Instagram in one instance showed a Delhi police officer kicking Muslim worshippers during namaaz (prayer), sanitising brutality through animation. While blurring fact and fiction, the image desensitises viewers to violence against minorities and further shapes public opinion.

Ghibli-style Islamophobic content

Some posts are seen using AI-generated images to suggest moral abnormality without explicitly saying so. The coded language and comedy hashtags serve as a shield, framing their intentions as harmless humour. 

“But the effect of these images, singly or cumulatively, is the dehumanisation of Muslim men as hypersexual predators and Muslim families as incestuous or morally corrupted units,” the report adds.

Khan emphasised the need for public awareness, stating that critical thinking is a requirement for people to learn to read these images and ask questions such as, “Who benefits from me believing this?”

The report indicates that the impact of this visual propaganda is stark, as it fuels a climate of fear, humiliation and exclusion for Indian Muslims.

Get the latest updates in Hyderabad City News, Technology, Entertainment, Sports, Politics and Top Stories on WhatsApp & Telegram by subscribing to our channels. You can also download our app for Android and iOS.