Experts are pointing to a significant shift in online harassment, particularly the usage of artificial intelligence to create sexualized images and propaganda specifically targeting Muslim women.
In New Delhi, a freelance model from Indian-administered Kashmir named Samreen Ayoub was taken aback upon discovering a troubling video circulating on Instagram that detailed her life in a fabricated narrative. This unsettling clip featured a voiceover, complete with made-up captions akin to a news segment, claiming she was a Muslim woman “selling her body” to Hindu men. Disturbingly, the video misidentified various individuals in her photos, even labeling her brother as her “pimp.”
“It felt like intense stalking,” remarked the 24-year-old Ayoub. “They had followed my life from my first semester at university through to the last.” She noted the presentation of the video felt so authentic that even close relatives could easily be deceived.
Ayoub’s experience is not an isolated incident. Researchers are seeing a growing trend where AI is increasingly used to create sexualized and harmful content targeting Muslim women. Many others affected by this phenomenon declined interviews for fear of shame and potential trauma.
The rise of such sexualized AI imagery of Muslim women coincides with India’s expanding involvement in global discussions about AI governance, highlighted by an AI Impact Summit earlier this year focused on regulatory issues.
A recent study from the Center for the Study of Organized Hate (CSOH), based in Washington, DC, examined 1,326 publicly available AI-generated images and videos across 297 accounts on social media platforms like X, Facebook, and Instagram from May 2023 to May 2025. The findings revealed that depictions of Muslim women attracted over 6.7 million engagements, making them the most interacted with category.
“Generative AI has rapidly turned sexual fantasies into striking imagery at minimal cost,” stated Zenith Khan, a coauthor of the study. This technology enables users, regardless of technical expertise, to transform derogatory narratives into misleadingly realistic visuals.
Organizations focusing on online safety, like Meri Trustline run by the RATI Foundation in Mumbai, have also reported an increase in cases involving digitally manipulated content. Their 2024 report indicated that around 10% of the 482 documented cases involved such material, a figure that is growing as AI tools become increasingly accessible and common.
“These violations remain largely silent due to the shame, fear, and trauma they invoke,” noted Salman Mujawar from the helpline, emphasizing that incidents often go unreported, even among close friends and family.
Following the release of Ayoub's video, her social media accounts became flooded with abusive comments, alongside threatening phone calls and damaging character accusations, creating what she described as a “digital lynching.” Subsequently, the CSOH's data set showcased AI-generated memes and fictitious pornographic images targeting Muslim women, often casting Muslim men in a negative light while portraying Muslim women as submissive.
Sahana Udupa, a media anthropologist, characterizes this trend as part of a "pornification of politics," where right-wing digital cultures utilize humor and sexualized imagery to normalize abuse against women and minority communities. This behavior perpetuates a hazardous ecosystem, drawing from deep-seated ideological roots beyond mere misogyny.
In this context, Muslim women's bodies are increasingly becoming battlegrounds for communal conflicts, as seen in the notorious cases of "Sulli Deals" and "Bulli Bai," mock auction platforms that specifically targeted Muslim women. These incidents highlight a concerning trend of exploiting women’s bodies for communal dominance, often with formal and informal backing from authorities.
As the prevalence of generative AI burgeons, the speed and scope of online harassment against Muslim women have intensified. Such tools are now freely available online, allowing users to generate sexualized images easily, amplifying the risk for those already experiencing abuse.
For activists like Afreen Fatima, the ramifications of such harassment extend into daily life. Since she openly criticized India’s Citizenship Amendment Act, she has faced growing online hostility and threats. The situation has shaped her perception of public safety, making her fearful when traveling alone due to the violent fantasies often projected onto Muslim women.
After Ayoub’s ordeal, her modeling career suffered as a result of the harassment. “If negative comments fill your profile, you won’t receive offers anymore,” she explained. Despite reporting the harassment to law enforcement, she experienced little to no action taken against the perpetrators.
Legal experts have pointed out that existing laws in India struggle to address the nuances introduced by AI-generated content. For instance, while penalties exist for sharing non-consensual images, those protections do not extend to images that are entirely fabricated by AI. Experts warn that without significant changes to laws and platform designs, the propagation of AI-generated abuse will only continue to proliferate unimpeded.
Ayoub expressed her desire to identify those responsible for the harassment, lamenting that she lost her reputation to individuals who had no understanding of who she was or the impact of their actions.
