Women in Pakistan are facing threats of violence and coercion after ordinary photographs are shared without their permission, while social media reporting systems sometimes reject complaints because the images contain no nudity or explicit sexual material. An investigation by The Guardian documented the case of a teenager whose selfie was used in a fake profile. She said the platform refused to remove it despite her fears for her safety.
The case reflects broader findings from Chayn, a nonprofit organization working with survivors of gender-based violence. Its research involving more than 60 Pakistani women describes how photographs that appear innocuous to automated moderation systems can be used for blackmail, intimidation and reputational attacks. Advocates are calling on technology companies to assess image-based abuse according to consent and the circumstances of the person depicted, rather than relying primarily on whether the image contains sexually explicit material.
Non-explicit photographs are being used to threaten women and evade moderation
The teenager interviewed by The Guardian described how a private photograph became a means of intimidation after a young man threatened to share it with her family. He created a social media profile using her image, and her complaint was rejected because the content was deemed compliant with the platform's rules. The resulting threat to her safety disrupted her education and left her dependent on family decisions about her future.
The circumstances illustrate a category of image-based abuse that can be difficult for platforms to recognize. A photograph does not need to be sexually explicit to carry serious consequences when it is circulated without permission or presented in a misleading context. In some families and communities, photographs showing women without a head covering or in clothing considered inappropriate can expose them to punishment, even when the image was originally taken in private.
Chayn's research into image-based abuse in Pakistan and the diaspora documents comparable experiences. One participant described how her husband accessed photographs stored on her phone, manipulated their presentation and circulated them among relatives. The resulting allegations damaged family relationships and interfered with her professional opportunities.
Another participant discovered that photographs taken at a wedding had been uploaded to a Facebook page featuring images of numerous young women without their permission. She reported the page and continued monitoring the comments, but the reporting process did not produce the removal she sought. In a separate account, a woman who had specifically asked friends not to share photographs showing her without her hijab found that the pictures had nevertheless been distributed.
These accounts describe different forms of conduct, including impersonation, unauthorized distribution and coercion. Their consequences depend partly on how the images are presented and who receives them. A conventional content review focused on visible nudity or sexual activity may therefore fail to recognize the alleged abuse, even when the person depicted reports an immediate risk to her safety.
Research identifies weaknesses in reporting systems and local-language moderation
Chayn's investigation, titled Explicit Harms of Non-Explicit Images, draws on interviews with more than 60 women from Pakistan and Pakistani diaspora communities. Participants described how images of everyday activities could be used to pressure them into unwanted relationships, expose them to family retaliation or undermine their access to education and employment.
The researchers found recurring difficulties with platform reporting procedures. Some women said they could not identify a complaint category that accurately described what had happened to them. Others submitted documentation or repeatedly contacted platforms without receiving an effective resolution. Chayn also identified cases in which moderation systems recognized abusive language in English but failed to respond to equivalent material written in Urdu.
The organization's policy analysis argues that established definitions of image-based abuse place considerable emphasis on nude or sexually explicit content. The researchers found that such definitions can exclude photographs whose harmful use depends on the absence of consent, the identity of the recipient or the threats accompanying their distribution.
The difficulty extends to formal complaints. Some participants reported that police or cybercrime authorities questioned their behavior before examining the alleged abuse, while others were told that intervention was unavailable because their photographs contained no explicit material. These experiences discouraged further reporting and sometimes increased the risk of unwanted disclosure.
Pakistan's Digital Rights Foundation has independently documented the volume of technology-facilitated harassment. Its latest annual helpline findings record 3,012 new complaints during 2025, including 2,586 cases categorized as cyber harassment. Women submitted 1,709 complaints, and the organization identified significant barriers to obtaining assistance in areas without accessible cybercrime investigation offices.
The helpline figures cover a broader range of online incidents and should not be interpreted as a count of non-explicit image abuse. They nevertheless provide context for the difficulties described by Chayn's participants, particularly the dependence on reporting systems that may require repeated explanations of sensitive personal circumstances.
Digital Rights Foundation founder Nighat Dad has argued that platform policies frequently overlook the ways abuse operates across different communities. The organization has called for more effective reporting procedures and moderation that accounts for local circumstances, including the potential for apparently ordinary images to be used to endanger women.
Advocates want platforms to recognize consent and assess threats in context
Chayn is urging social media companies to revise how they classify image-based abuse, giving greater weight to whether the person depicted authorized the image's distribution. Its recommendations include introducing reporting procedures specifically designed for non-consensual sharing and allowing users to explain circumstances that may make otherwise ordinary content dangerous.
The organization also recommends investment in moderation teams with relevant linguistic and cultural expertise. Its research identifies the limitations of automated classification when a photograph's significance depends on local social expectations or accompanying intimidation. A review process that recognizes the image as harmless in isolation may reach a different conclusion if it considers evidence of impersonation or threats made to the person depicted.
Other proposals address the handling of complaints after they are submitted. Chayn recommends allowing trusted third parties to report abuse on behalf of affected individuals, particularly when the person targeted cannot access the material being circulated. It also advocates temporary restrictions on reported images during review and mechanisms for preserving evidence when content is removed.
Such changes would require platforms to reconcile requests for removal with other legitimate uses of photographs, including public-interest reporting and freedom of expression. Chayn's proposed approach gives substantial weight to individual consent, while its policy recommendations acknowledge that any broader legal right to control identifiable images would need appropriate exceptions.
The Guardian reported that Meta did not respond to its request for comment. The article did not establish whether the company operated the platform involved in the teenager's rejected complaint, and the available evidence does not provide a comprehensive account of moderation decisions across individual companies. The documented cases therefore support criticism of specific reporting experiences without establishing the frequency of comparable failures across entire platforms.
The research also shows that removal alone may be insufficient once an image has been copied or circulated through private messaging channels. The threat of further distribution can continue even after the original post disappears, while victims may need records of the material to pursue legal remedies. Chayn's recommendations address these problems by connecting content removal with evidence preservation and more accessible complaint procedures.
The findings leave technology companies facing questions about whether their existing systems can identify image-based abuse whose severity is determined by context rather than visual content. For the women documented in the research, the consequences of inadequate enforcement have included lost educational and employment opportunities, family retaliation and prolonged exposure to threats. The organizations are seeking changes that would allow those risks to be considered during moderation before further harm occurs.