By Syeda Aliza Sajjad

1. The Zombie Bite: Integration of GenAI in all its Violent Glory

To exist as a woman in Pakistan is nothing but a game of Russian roulette. What kind of violence or harassment will I suffer today? The avenues are endless—on the street, in your homes, schools, universities, offices, and now, with the advent of technology, in digital and virtual spaces too. Violence simply shapeshifts; it evolves with society. And today, that evolution has taken the form of generative AI (GenAI).

Technology, social media, and digital platforms, when left unregulated, are increasingly becoming tools of violence, oppression, and harassment. This isn’t news by any means. The seamless uptake and integration of generative AI is, though. Women are disproportionately targeted by every new technology that encroaches on our lives like a leech. GenAI, more commonly understood and accepted as ChatGPT, is the Zombie bite in an apocalyptic world that will slowly strip us of our remaining humanity.

In Pakistan, ChatGPT, given a “nickname” of GPT as is the norm with people we’re close to, has grown to become one of the most used applications, with over six million users. But who’s keeping a check on what this adoption into our daily lives looks like, and who’s bearing the brunt of this careless adoption?

Recently, there has been an uptick in generating pictures of women, either to dress them more “modestly” if they're “showing skin”, or undressing them (“AI Nudification”), another way to dehumanize and sexually assault women and girls on social media. The realistic nature of GenAI content makes the misuse all that sinister.

This uptick cannot be viewed separately from the broader context of patriarchal control over a woman’s body in online and offline spaces; it is a manifestation of the same entitlement and control women witness in other spaces. There is a sheer lack of knowledge, awareness, and understanding around GenAI and its harms, with the open-access technology being rapidly adopted across industries and platforms and little regard given to how it’s being used. Pakistan being one of the top users of ChatGPT, and at the bottom in gender equality, puts Pakistani women, girls, and trans people in an incredibly vulnerable position

2. A New Tool for an Old Crime

GenAI generates artificial images, videos, voices, and text in response to prompts. The content appears “novel,” yet is stitched together from existing online data. Crucially, these platforms are free and open to all. With zero technical skill required, anyone can create hyper-realistic content with the click of a button.

The advent of new avenues, like GenAI, found men getting creative (read: typical!) with the different ways they can and will harass women. AI has opened the floodgates for something far more sinister than ever before. Deepfakes are increasingly becoming indistinguishable from reality with how GenAI creates hyper-realistic images and videos with the click of a button. Where perpetrators had to put in the effort of using Photoshop, their work still sloppy, GenAI has removed any remaining barriers with how accessible it is for an average smartphone user to put any image of a woman through the platform and fabricate it however they like.

Deepfakes have become nearly indistinguishable from reality. According to global research, 98% of deepfake videos are pornographic, and 99% of those target women and girls. The numbers are horrifying but unsurprising.

In Pakistan, women journalists, politicians, and activists have increasingly been targeted with deepfake videos meant to humiliate, silence, and punish them. GenAI creates ready-made templates for harassment, misinformation campaigns, and fabricated personal histories.

3. Patterns of yesterday, today, and tomorrow: AI-Facilitated Sexual Violence

With the normalization of GenAI on social media platforms like Instagram, TikTok, Facebook, and X (formerly known as Twitter), where users are generating seemingly harmless images of themselves in “ghibli-themed” animations, hugging their favourite celebrities, and converting decades old pictures to short videos, there is a disturbing rise in the way men interact with women and girls. Women and girls on X, across the world, have noticed men sharing images of themselves through GenAI to either “undress” or “modestly cover” them, both demonstrating a form of sexual harassment that’s indicative of the patriarchal, entitled mindset that reduces women’s bodies, choices, and existence to mere objects to play with. The violence is intimate yet public, meant to degrade and humiliate women and girls into silence.

It's a fascinating phenomenon if one digs deeper into it. The “AI Nudification” of women’s photos online is not new in nature, but new in execution. With every new “AI Trend” that plagues our social media, the technology is being trained to execute fake images with precision. Where women who dare to make their presence known and felt in online spaces have been policed by men on what they are (or aren’t) wearing, GenAI has given them the magic wand, paired with their existing audacity, to act on those entitled thoughts. Social media is seeing a rise in replies by unknown, anonymous men to women's posts of themselves. Depending on the audacity of the man, she’s either “too covered”, and he’ll strip her, or “too vulgar” and he’ll cloak her. In any and all cases, a man’s unsolicited presence is known and felt in a woman’s space, despite it being virtual in nature. The same misogynistic attitudes that lead to street harassment or domestic violence manifest in online spaces, with men feeling entitled to alter, control, or "correct" how women present themselves digitally. Digital violence, like AI nudification and adjacent harassment, reflects and reinforces traditional patriarchal norms that women are burdened with on a day-to-day basis, now extending to the digital realm in a new, much more rigorous format.

4. Pakistan’s Patriarchal Positionality in Online and Offline Spaces

Pakistan is already a high-risk environment for women online:

● A 2018 Digital Rights Foundation survey found 72% of Pakistani women had experienced online harassment, including unwanted messages, threats, and stalking.
● COVID-19 lockdowns further increased online harassment and abuse, isolating women and leaving them more vulnerable.
● Conservative norms mean families often blame women for harassment, silencing victims before they can seek help.

In such a context, the rise of GenAI is catastrophic. AI nudification and deepfakes become a digital extension of societal beliefs that women are responsible for their own victimization.

Women in Pakistan carry the trauma of harassment in their bodies and daily lives. Using AI to sexually harass women is meant to intimidate them out of public spaces, to reinforce the marginalization of women and girls from digital spaces, and to strip them of access to employment and educational opportunities. In the era of digitalization in a conservative society where women are blamed for the actions of perpetrators and treated as second-class citizens, making women feel unsafe, uncomfortable, and fearful of interacting in online spaces is meant to exclude them from public spaces, sociopolitical discourses, and the benefits of digital technology, perpetuating a greater digital gender divide.

AI-facilitated sexual harassment causes immense distress to women and girls by spreading disinformation, with many families viewing it as a punishable offense and many women withdrawing from online communities altogether. The harms are also psychological and mental; women are made to feel shame and humiliation at the actions of perpetrators, the blame wrongfully resting on their shoulders. They suffer through PTSD, anxiety, and depression in fighting against AI-facilitated sexual harassment and seeking justice.

The entitlement men feel toward women’s bodies offline seamlessly transfers online. The same mindset that fuels street harassment, domestic abuse, and sexual violence now fuels AI-generated violations. Men feel entitled not only to comment on a woman’s appearance but to alter it, manipulate it, “correct” it.

In Pakistan’s context, where accusations about a woman’s “character” can derail her education, career, and physical safety, GenAI becomes an especially potent tool of control.

5. Vulnerability and Intersectionality: “He said, He said”

Not all women face digital violence equally. Class, geography, age, and gender identity shape vulnerability.

Women from lower-income households and rural communities often lack digital literacy or access to safety tools. Many are already restricted by families or male relatives from using mobile phones or the internet. When they do access technology, harassment or deepfakes can lead to intense backlash, including physical violence.

The Khwajasira and transgender community in Pakistan face an even greater risk. Already marginalized, policed, and attacked in physical spaces, online deepfakes and harassment multiply their vulnerability. They have been subjected to coordinated digital attacks and smear campaigns: pre-transition images and national identity cards revealing their personal information and deadnames have been leaked, and they are threatened with the release of non-consensual intimate videos, bringing an onslaught of violence and hate to the community. Manzil Foundation, a grassroots organization working on transgender rights in Khyber Pakhtunkhwa (KP), recorded seven cases of TFGBV against Khwajasiras and transgender persons towards the end of 2024, with at least two Khwajasira individuals having been killed by anti-trans groups for resisting the threats and harassment.

Younger girls are particularly at risk. They are more likely to use technology for education, to access learning opportunities, and to connect with peers on social media and messaging platforms. When this increased online presence is combined with the general lack of media and digital literacy and safety in Pakistan and in educational spaces--and with social media and digital spaces rife with predators and groomers--girls often become easy targets of sexual violence. A global study reported that 58 per cent of girls and young women have experienced some form of online harassment. Their images, often innocent and shared within closed circles, can be taken, manipulated, and circulated widely. They are also dependent on the adults in their lives and may find it difficult to report abuse or come forward, fearing that they might lose access to technology, face blame, or encounter further restrictions.

The long-term consequences are devastating: shame, isolation, ruined reputations, mental health struggles, and reduced access to education or employment.

When digital spaces become unsafe, women withdraw. This withdrawal widens Pakistan’s digital gender divide, reproducing offline inequalities in an online world that should have offered liberation.

There is also a lack of reporting, investigation, and research on the growing use of generative AI to sexually harass women, and this gap in information discourages deeper examination of the issue. Existing patterns of technology-facilitated gender-based violence (TFGBV) reveal who is most targeted and the consequences they face, but the novel and uncharted nature of generative AI leaves many questions about its use, spread, and societal integration unanswered.

On November 16, 2025, Benazir Shah, Editor of Geo Fact Check, shared a post on X exposing an AI-generated deepfake video showing her “dancing in a bar in London.” She immediately called out the account using the hashtag #AttacksWontSilenceUs and urged Information Minister Tarar Attaullah to take notice. Her case attracted media attention, and the Information Minister responded positively. Yet, the video remains on X, despite being reported for targeted harassment. While X allows AI-generated content on the platform, it also has broad policies against impersonation and the creation of fake personas using AI. But despite the language in its guidelines, enforcement is far from consistent. Harmful fake content and disinformation often remain online for too long. Posts must be reported numerous times before the platform evaluates them, and even then, the case can ultimately be dismissed. X has clearer guidelines on child safety, but none that protect girls and women from cyber harassment in particular. What does this reveal about the effectiveness of existing channels for reporting violence, or about how seriously sexual harassment is taken in an era of growing digital threats?

Her case is not isolated. It is part of a disturbing pattern of digital attacks targeting high-profile women to discredit, intimidate, and force them out of public life.

Journalists, activists, and politicians have repeatedly reported AI-generated sexualized images circulating online. The intention is clear: weaponize technology to shame women, undermine their credibility, and attack political opposition through their bodies.

These cases reveal a grim truth: AI deepfakes are a form of political violence as much as sexual violence.

6. Accountability and Regulation by the State and Big Tech

Platforms like X, Instagram, TikTok, and Facebook have consistently failed to detect or regulate AI-generated content. Reporting mechanisms are slow, opaque, and ineffective. Deepfakes spread in minutes, while takedowns take days, if they happen at all.

Tech companies have the capacity to build detection tools, but they lack the urgency and political will to protect women.

Pakistan has legal frameworks that should, in theory, address these harms:

● The Prevention of Electronic Crimes Act (PECA) 2016 criminalizes cyberstalking, non-consensual sharing of intimate images, identity misuse, and child pornography.
● The Protection Against Harassment of Women at the Workplace Act includes digital harassment.
● Pakistan Penal Code provisions cover defamation, intimidation, and obscenity.

But the gap between law and implementation is enormous. Law enforcement often lacks the technical skills to investigate AI-generated crimes. Victims face disbelief, dismissal, or moral policing when reporting cases. Patriarchal norms discourage women from seeking justice at all.

A strong legal framework is meaningless if women cannot safely access it. At the same time, we must conduct our own research to understand what integrating generative AI into our lives might unleash. This technology is as dangerous as it is unchecked, unregulated, and uncritical. Platforms like ChatGPT impose virtually no limits on what can be asked of them, leaving the door wide open to venture into uncharted territories in ways that may be difficult to undo.

7. Women Will Always Fight Back

Despite everything, women are not silent. Pakistani digital rights groups, feminist organizations, and activists continue to push for safer digital environments, stronger laws, and better enforcement.

Campaigns like #AttacksWontSilenceUs and initiatives by organizations such as the Digital Rights Foundation have built support networks for survivors, offering legal aid, digital security training, and public advocacy.

Globally, women are also experimenting with AI tools themselves to build safer online ecosystems, document abuses, and demand accountability from Big Tech. The rise of generative AI has created a new frontier for old patterns of violence. Deepfakes, AI nudification, and digital manipulation are not only tech problems but rather patriarchal problems wearing new skins.

The urgency cannot be overstated: Governments must regulate AI while tech companies must build tools to detect and remove deepfakes, and we must hold companies and perpetrators accountable while continuing to educate and advocate.

The question becomes: what will it take for women and girls to be treated with dignity, respect, and humanity in every virtual and physical space, and when will Pakistan and its citizens become responsible users of digital platforms and technology?