Posts Condemning Bus Harassment

SUMMARY

Meta should expand policy exceptions and provide greater clarity for reviewers regarding awareness-raising content, the Oversight Board has found in assessing two cases involving videos of sexual harassment on public transportation in India and Brazil. These actions are particularly important given concerns about monetization and engagement-baiting that employ purported awareness-raising content.

Meta needs to bridge the gap between external policy and internal guidance on “creepshots,” non-consensual imagery taken of unsuspecting people, so reviewers can effectively moderate awareness-raising content.

The Board finds that neither post violated Meta’s community standards because they are awareness-raising content, and that removing them was inconsistent with the company’s human rights responsibilities. They should have received exceptions.

The Board also concludes that by making both posts unavailable for recommendation to other users, Meta is disregarding how stakeholders use its platforms for advocacy, including reaching policymakers and lawmakers, and compromising the reason for relevant policy exceptions.

Why This Matters

A 2012 survey of 143 countries found that three in five women worldwide experience sexual harassment on public transportation. Violence against women remains one of the “most … under-addressed human rights crises” globally, according to the World Health Organization. Meanwhile, social media is a powerful tool for advocacy and awareness-raising for social causes, particularly gender-based violence, and for survivors to share experiences, which is especially important where there is mistrust in formal reporting systems.

These two cases depict sexual harassment on public transportation in India and Brazil, where previous instances of sexual violence on public transportation have gained national attention. Through these cases, the Board is assessing Meta’s moderation practices when users intend to raise awareness about sexual harassment and gender-based violence.

About the Case

The first case involves a minute-long video posted to a prominent Indian news organization’s Facebook page in November 2025 in which an unidentifiable woman films a man sitting next to her on a bus in India. The man attempts to grope her by reaching under her shirt, and she verbally confronts him. The caption raises awareness about women’s safety.

A user reported the content on the day it was posted. Human reviewers and policy subject-matter experts determined it violated the Adult Sexual Exploitation Community Standard and removed it.

The second case involves an 18-second video from Brazil posted to Instagram in December 2025. The video – recorded by an unidentifiable woman sitting next to a man on a bus – shows the man holding his clothed genitalia, while he looks at his phone. The video’s text overlay and the caption provide the incident’s location and highlight concern over women’s safety and harassment on public transportation.

Two users reported the content on the day it was posted. After human review, it was removed under the Adult Sexual Exploitation policy for “creepshots.”

The posting users failed in their appeals against the respective decisions.

Following the Board selecting these cases, Meta determined its initial decisions were wrong and restored both posts with age-gating and a warning screen applied.

Key Findings

The Board finds that neither the India nor the Brazil case violates Meta’s community standards because they fall under policy exceptions for awareness-raising and body contours (which allows content showing the outline or shape of breasts, buttocks, or genitals through clothing or other coverings, if no other nudity is visible), respectively. Both should have remained on the platform with appropriate age restrictions and warning screens. These cases raise concerns about enforcement accuracy, especially considering that multiple human reviewers assessed both pieces of content.

Considering research that views public masturbation as a form of sexual harassment, the Board encourages Meta to expand its definition of sexual harassment under the Adult Sexual Exploitation policy awareness-raising exception beyond “non-consensual touching” to include other forms of sexual harassment.

The Board finds that removing both pieces of content was not consistent with Meta’s human rights responsibilities. Both posts serve awareness-raising purposes. Neither revealed the survivors’ identities, and each contained clear signals in their captions that they were raising awareness of sexual harassment and violence. Warning screens with age restrictions were legitimate restrictions. Warning screens allow users to have critical conversations on violence and safety while alerting other users to the discussion’s potential sensitivities. An age restriction protects minors from injurious material.

Coupling the “Mark as Sensitive” warning screens with the non-recommendable function was not a legitimate restriction on freedom of expression. The Board is concerned that by making both posts non-recommendable, Meta is ignoring the value of its platforms for survivors, advocates and intended audiences, including legislators – ultimately undermining the purpose of the relevant awareness-raising policy exception.

The Adult Nudity and Sexual Activity policy rationale lacks a clear and operational criteria for an awareness-raising exception. This must be codified for users and moderators. The Adult Sexual Exploitation policy is also insufficiently clear. It allows depiction of non-consensual sexual touching for awareness-raising, provided there is no “sensational context,” yet does not define “sensational context.”

The Board is also concerned about the disconnect between Meta’s external policies and the enforcement decision in the Brazil case. The Board agrees with Meta’s conclusion that the content did not violate the Adult Sexual Exploitation’s “creepshot” policy because it did not intend to humiliate or mock the individual filmed. However, internal guidance definitively states that any content depicting masturbation in public and which mocks, sexualizes or identifies an individual – including content intending to condemn the behavior – violates the “creepshots” policy, meaning reviewers had to remove it. At the same time, the rationale in the public-facing policy recognizes the importance for “survivors to share their experiences.” These internally and externally facing factors are not consistent and create confusion for content reviewers and users.

The Oversight Board’s Decision

The Board overturns Meta's decision to remove the content, which should have been left up with the appropriate age-gating and warning-screen restrictions. Neither piece of content should have been made non-recommendable.

The Board recommends that Meta:

• Expand the Adult Sexual Exploitation policy exception language from “content depicting non-consensual sexual touching” to “content depicting non-consensual sexual touching and/or other forms of sexual violence.”
• Include an exception for awareness-raising under the Adult Nudity and Sexual Activity policy, with the same consequences as the Adult Sexual Exploitation policy exception, including age-gating and warning screens.
• Align its internal guidance and external language on the “creepshots” policy line of the Adult Sexual Exploitation Community Standard to clarify that not all depictions of public masturbation are violative.
• Should decouple the “Mark as Sensitive” warning screen from the non-recommendable enforcement action.

  *Case summaries provide an overview of cases and do not have precedential value.

Full Case Decision

1. Case Description and Background

This decision highlights two cases depicting instances of sexual harassment on public transportation presented with the aim of raising awareness. According to UN Women, the lead United Nations entity on gender equality, nearly 90% of women and girls experience some form of violence when accessing public transportation. The World Health Organization’s Violence Against Women Prevalence Estimates 2023 Report concludes that violence against women remains one of the “most persistent and under-addressed human rights crises” globally, with very little progress recorded since 2000. The report states that nearly 840 million women - or one in three women – have experienced sexual violence in their lifetime.

Social media is regularly utilized as a platform to raise awareness on social issues such as sexual exploitation or violence. A World Bank study found that educational content delivered through social media can change attitudes towards violence against women. Beyond being an advocacy tool, social media is often a resource for survivor documentation. Research shows that a sexual harassment history was the most influential predictor of engaging with the social media campaign #MeToo. Yet, awareness-raising content also complicates gaps in content moderation as it reveals the limited ability of platforms to distinguish between content that perpetuates the harm versus content that exposes or condemns the harm. As previous Board decisions have stated, distinguishing between violative content and awareness-raising content requires clear policies and nuanced enforcement (see India Sexual Harassment Video).

The first case involves a minute-long video posted to a Facebook page in November 2025 in which a woman, who is not identifiable in the recording, films a man sitting next to her on a bus in India. The man attempts to grope her by reaching under her shirt and she verbally confronts him. The accompanying caption praises the victim’s reaction, expresses dismay at the lack of support shown by those around her and raises awareness about the issue of women’s safety. The Facebook page belongs to a prominent news organization in India with more than 1.1 million followers. The case content was viewed more than 60 times before Meta took it down.

A user reported the content on the same day it was posted for Adult Sexual Exploitation. Human reviewers determined it to be violative of this policy. The case content was then sent for additional review, and afterwards to policy subject matter experts who all determined the content to be in violation of the policy. The post was removed in December 2025. As the content was taken down for non-consensual sexual touching, the case content creator received a strike and a time-based feature limit. Non-consensual sexual touching is any form of sexual activity or sexual touching (as defined by the Adult Nudity and Sexual Activity policy) that occurs without consent. Lack of consent is determined by context. The content creator appealed, and upon human review, the company maintained its decision that the content was violative.

The second case involves an 18-second video from Brazil posted to Instagram in December 2025. The video, also recorded by an unidentifiable woman sitting next to a man on a bus, shows the man holding his clothed genitalia, with his side profile visible while he is looking at his phone. The text overlay on the video and the caption provide details about the incident – including the bus route and stop where the individual got off the bus – and expresses concern for women’s safety and the issue of harassment on public transportation. The posting user self-identifies as a journalist and their bio encourages viewers to advertise on their page. The content had more than 150,000 views before it was removed.

Two users reported the content on the same day it was posted – once for prostitution, under the Adult Sexual Solicitation and Sexually Explicit Language policy, and another for Adult Nudity and Sexual Activity. After human review, the post was removed under the Adult Sexual Exploitation policy for creepshots, and consequently a standard strike was applied to the case content creator’s profile. The content creator appealed and, upon human review, the company maintained its decision that the content was violative.

In each case, the case content creator appealed to the Board.

As a result of the Board selecting these cases, Meta determined its initial decisions were wrong and restored both posts to their platforms with age-gating and a warning screen applied.

Both cases resurface previous incidents of sexual crimes on public transportation that gained national attention in their respective countries. India’s Nirbhaya case of 2012, a horrific sexual crime which occurred on a bus, shocked the country and led to new legislation to address sexual violence. In Brazil, a 2017 case involving public masturbation on a bus sparked outrage on social media and raised calls for women’s safety in public spaces. The topics depicted in the cases discussed in this decision are not isolated and reflect a broader issue around women’s safety in public spaces in India, Brazil and globally.

Despite new legislation in India since 2012, crime against women remains prevalent. According to India’s National Crime Records Bureau, 448,211 cases of crimes against women were recorded in 2023 (up 0.7% from 2022), equivalent to roughly 51 complaints per hour. These crimes include rape, stalking and sexual harassment. Experts retained by the Board explained that low rates of confidence in police and legal systems may lead to underreporting and inadequate redress. As the United Nations (UN) Special Rapporteur on Sexual Violence noted upon their country visit to India, “sexual violence, including rape and sexual harassment, is widespread across the country and perpetrated in public and private spaces” (para. 7). The World Bank reported in 2021 that 56% of women who used public transport across metropolitan areas in India reported being sexually harassed.

Official reports show that gender-based violence is on the rise in Brazil, with more than a 6% increase in sexual harassment cases from 2023 to 2024. A 2025 report said that 37.5% of women aged 16 and over had experienced some form of violence during the previous 12 months. In 2019, a survey found that 97% of Brazilian women aged 18 and over who used public or app-based transportation reported experiencing some form of harassment in transit. The country has implemented legal and policy changes to address these issues, including the criminalization of public sexual harassment in 2018.

Social media has become a powerful tool for advocacy and awareness-raising for social causes, particularly around gender-based violence. The aforementioned legislative changes of 2018 in Brazil were – at least partially - a result of the hashtag campaigns online that spoke out against sexual violence. As UN Women reports, online movements like #MeToo showcase “exhaustion and anger about sexual harassment across the world, and hunger for urgent change.” Social media also serves as a tool for survivors to share their stories and experiences, which is particularly important in settings where trust in formal reporting systems is lacking.

Monetization and engagement may incentivize creators to share emotionally engaging (including violent) content online. Research indicates that emotionally engaging content, including that which evokes shock, anger or concern, tends to generate stronger engagement and may be more widely distributed on platforms. As experts to the Board explained, these systems are rarely able to distinguish between engagement driven by concern, empathy or public interest and engagement driven by curiosity, outrage or other motivations. Thus, controversial content – such as depictions of gender-based violence – can potentially lead to an increase in engagement and followers which can then drive revenue.

2. User Submissions

In a statement to the Board, the user who authored the content in the India case explained that the purpose of the post was to serve the public interest and “not to glorify violence or shame anyone, but to ... start an urgent conversation about women’s safety, bystander responsibility and collective accountability.” They also said that “suppressing such content risks silencing the very voices that need to be heard.”

The content creator of the Brazil case similarly explained to the Board that the aim of the video was to seek “prevention, awareness and safety, especially for women who face this type of situation on a daily basis.” The user states that removing the content silences a victim, in doing so “preventing important information from reaching the community.”

3. Meta’s Content Policies and Submissions

I. Meta’s Content Policies

The Adult Sexual Exploitation Community Standard recognizes Meta's platforms as “a place to discuss and draw attention to sexual violence and exploitation” and that this is an “important part of building common understanding and community.”

Under this policy, Meta removes imagery depicting non-consensual sexual touching (except in real-world art depicting non-real people, with a condemning or neutral caption). It also removes “descriptions of non-consensual sexual touching, unless shared by or in support of the survivor.”

So-called “creepshots” or “sharing secretly taken non-commercial imagery focusing on a person's commonly sexualized body parts (such as breasts, groin, buttocks or thighs)” with the intent to mock, sexualize or reveal the identity of the person depicted in the imagery are also removed.

Meta may, as outlined in the policy exception of the Adult Sexual Exploitation Community Standard, include a sensitivity screen for “narratives and statements that contain a description of non-consensual sexual touching (written or verbal) that includes details beyond mere naming or mentioning the act if [...] shared by a third party in support of the survivor or condemnation of the act or for general awareness to be determined by context/caption.”

Moreover, the policy exception states that Meta may restrict visibility to people over the age of 18 and include a warning label on “certain content depicting non-consensual sexual touching, when it is shared to raise awareness and without entertainment or sensational context, where the victim or survivor is not identifiable and where the content does not involve nudity.”

The policy rationale for the Adult Nudity and Sexual Activity Community Standard states that the policy removes “real photographs and videos of nudity and sexual activity, AI- or computer-generated images of nudity and sexual activity, and digital imagery, regardless of whether it looks ‘photorealistic’” though it makes careful allowances for certain “awareness-raising content.” It also removes photorealistic/digital videos that focus on crotch, female breasts or buttocks recorded without the awareness of the person(s) depicted in them.

The Adult Nudity and Sexual Activity Community Standard restricts the visibility of content that is “photorealistic/digital imagery and real-world art of explicit, implicit or other sexual activity or stimulation when only body shapes or contours are visible” to people over the age of 18. The policy also allows a label on certain content including “photorealistic/digital imagery and real-world art of visible genitalia" so users know the content may be sensitive.

II. Meta’s Submissions

Meta told the Board that when the case content was escalated to an internal reviewer with regional expertise, they concluded that the contextual criteria for the exception were not met.

After the Board selected the India case, Meta reversed its decision and restored the content – with restricted visibility and a warning screen – because all conditions for the escalation-only Adult Sexual Exploitation policy exception were met – the content depicted non-consensual sexual touching without entertainment or sensational context and was posted to raise awareness with no nudity and the survivor unidentifiable.

When restoring the India case content to the platform, Meta added a “Mark as Sensitive” warning screen. When content receives a “Mark as Sensitive” warning screen, it automatically becomes non-recommendable.

In the Brazil case, following the Board’s selection, Meta reversed its decision and restored the content with restricted visibility. While the content was originally removed under the “creepshots” provision of the Adult Sexual Exploitation policy, Meta later determined it was outside the scope of this policy line because it was not shared with an intent to mock, sexualize or identify the individual. Instead, Meta submitted that it was more appropriately enforced under the Adult Nudity and Sexual Activity policy. According to Meta, while the video was determined to show self-stimulation, the individual is clothed and only body shapes and contours are visible, thus the content did not violate the Adult Nudity and Sexual Activity policy.

According to Meta’s internal guidance, sharing imagery of someone masturbating in public to condemn the act violates the Adult Sexual Exploitation policy if, in addition to condemnation, the content mocks, sexualizes, or reveals the identity of the person depicted in the image, per the “creepshots” policy line.

While it is not generally applied under the “body contours” policy line of the Adult Nudity and Sexual Activity policy, Meta applied a “Mark as Sensitive” warning screen to the content in the Brazil case on escalation because of its increased sensitivity. The warning screens can be applied through automated systems or by human reviewers. By applying the screen, the content was made non-recommendable.

The Board asked questions on, among others, the Adult Sexual Exploitation Community Standard, the Adult Nudity and Sexual Activity Community Standard, and enforcement mechanisms such as age-gating and warning screens. Meta responded to all the questions.

4. Public Comments

The Board did not receive any public comments that met the terms for submission.

5. Oversight Board Analysis

The Board selected this case to analyze Meta’s policies and enforcement practices related to awareness-raising content on issues of sexual violence and women’s safety. The case falls within the Board’s Gender strategic priority.

The Board analyzed Meta’s decision in this case against Meta’s content policies, values and human rights responsibilities. The Board also assessed the implications of this case for Meta’s broader approach to content governance.

5.1 Compliance With Meta’s Content Policies

I. Content Rules

India Case

The Board finds that the India case does not violate Meta’s Adult Sexual Exploitation Community Standard because it falls directly under the policy exception for awareness-raising content. As such, it should have been allowed to remain on the platform with appropriate age restrictions and a warning screen. In this case, all three conditions for awareness-raising content depicting non-consensual sexual touching were met: (1) it was shared to raise awareness without entertainment or sensationalist context; (2) the survivor is not identifiable; and (3) the content does not involve nudity. This is aligned with the Adult Sexual Exploitation policy exception.

Brazil Case

The Board finds that the Brazil case does not violate Meta’s Adult Nudity and Sexual Activity Community Standard because it falls directly under the body contours exception. The Board agrees with Meta’s conclusion that this case was best enforced under this exception. As the individual is fully clothed, the content depicts stimulation where only body contours are visible. As such, it should have been allowed to remain on the platform but should have received the appropriate age restrictions. Due to the sensitive and sexualized nature of the content, it should have also received a warning screen to properly inform users of the content type.

This content does not fall under the Adult Sexual Exploitation policy line on “creepshots” as it was not shared with the intent to mock, sexualize or reveal the identity of the individual. There are no attempts to show the identifying facial features of the individual or to share personal identifying information about them. It also does not fall under the awareness-raising policy exception either because it does not depict non-consensual sexual touching. As it is content depicting public masturbation, the Board notes that had it revealed the individual’s identity, even if it was condemning the act, it would have been taken down under the Adult Sexual Exploitation policy’s internal guidance. However, that internal enforcement action is unclear in the public language of the policy.

While this particular case is better enforced under the Adult Nudity and Sexual Activity Community Standard, the Board is concerned that the policy exception under Adult Sexual Exploitation is too narrow to properly engage with content depicting all forms of sexual violence. As the policy rationale states, Meta’s services play an important role as “a place to discuss and draw attention to sexual violence.” Public masturbation is considered a form of sexual harassment by some, who point out that the “unwanted nature of the encounter implies sexualization, intimidation and discomfort” which is evidence of sexual harassment. UN Women defines sexual violence as “any sort of harmful or unwanted sexual behavior that is imposed on someone.” The Board encourages Meta to expand the Adult Sexual Exploitation awareness-raising policy exception beyond “non-consensual touching” to include other forms of sexual violence or harassment. Any potential awareness-raising policy exception should be properly developed to account for content that condemns public masturbation.

The Board also recognizes that the depiction of sexual violence online impacts individuals differently across the globe due to differing cultural contexts and nuances. The norms of public sharing about these topics vary widely and reasonable differences of opinion are possible. Awareness-raising content is often transgressive in nature as it often is in tension with local norms. As such, the Board recommends erring on the side of speech. Meta’s efforts to consider context in each of these cases – as demonstrated in their deployment of regional expertise in human review - should be continued in similar cases depicting sexual violence, including when shared to condemn the behavior. However, as the outcome of the reviews in these cases show, human reviewers need better guidance to properly assess both the policy lines and the specific contexts in which these can be applied.

II. Enforcement Action

These cases raise concerns about accuracy in enforcing the Adult Sexual Exploitation and the Adult Nudity and Sexual Activity Community Standards, especially considering that multiple reviewers assessed both pieces of content.

Regarding the India case, the Board is concerned that different human reviewers analyzed the post at various points and they all failed to properly assess it as permitted under the awareness-raising policy exception. Due to the sensitive nature of this content and the fact that the policy exception is only applied on escalation, human reviewers need to be able to properly assess context and successfully recognize when the content fits the exception. This case was reviewed by human reviewers three times. It is particularly concerning to the Board that even though this resulted in an additional escalation to policy subject matter experts, they still assessed the content as violative and did not capture it as a policy exception.

5.2 Compliance With Meta’s Human Rights Responsibilities

The Board finds that removing both pieces of content from the platform was not consistent with Meta’s human rights responsibilities. The use of warning screens and age restrictions were legitimate restrictions on both pieces of content. However, the non-recommendable outcome of the warning screens was not a legitimate restriction.

Freedom of Expression (Article 19 ICCPR)

Article 19 of the ICCPR (International Covenant on Civil and Political Rights) provides for broad protection of expression, including expression that may be considered “deeply offensive” (General Comment 34, para. 11).

When restrictions on expression are imposed by a state, they must meet the requirements of legality, legitimate aim, and necessity and proportionality (Article 19, para. 3, ICCPR). These requirements are often referred to as the “three-part test.” The Board uses this framework to interpret Meta’s human rights responsibilities in line with the UN Guiding Principles on Business and Human Rights, which Meta itself has committed to in its Corporate Human Rights Policy. The Board does this both in relation to the individual content decision under review and what this says about Meta’s broader approach to content governance. As the UN Special Rapporteur on freedom of expression has stated, although “companies do not have the obligations of governments, their impact is of a sort that requires them to assess the same kind of questions about protecting their users' right to freedom of expression” ( A/74/486, para. 41).

I. Legality (Clarity and Accessibility of the Rules)

The principle of legality requires rules limiting expression to be accessible and clear, formulated with sufficient precision to enable an individual to regulate their conduct accordingly (General Comment No. 34, para. 25). Additionally, these rules “may not confer unfettered discretion for the restriction of freedom of expression on those charged with [their] execution” and must “provide sufficient guidance to those charged with their execution to enable them to ascertain what sorts of expression are properly restricted and what sorts are not” (ibid). The UN Special Rapporteur on freedom of expression has stated that when applied to private actors’ governance of online speech, rules should be clear and specific (A/HRC/38/35, para. 46). People using Meta’s platforms should be able to access and understand the rules and content reviewers should have clear guidance regarding their enforcement.

While the Board welcomes Meta’s implementation of the awareness-raising policy exception in the Adult Sexual Exploitation Community Standard (a result of the recommendations from the India Sexual Harassment Video decision), it is concerned that the Adult Nudity and Sexual Activity Community Standard does not provide similar clarity to users. Rather, its policy rationale makes only a passing reference to awareness-raising without setting out clear and operational criteria and details as to what is permitted on the platform. The policy rationale must be codified in the operable rules of the policy to avoid confusion for users and content moderators. In its responses to the Board, Meta justifies the lack of a freestanding exception for awareness-raising under the Adult Nudity and Sexual Activity policy by stating that content can be reviewed on escalation by policy subject matter experts who may decide to give it an exception such as a newsworthiness allowance.The Board does not believe that this is sufficient as the allowance can be, as Meta states, narrow in nature. Meta should provide more details to users and to those enforcing the rules about what awareness-raising content looks like under the Adult Nudity and Sexual Activity policy.

The exception for awareness-raising under the Adult Sexual Exploitation policy specifically allows content depicting non-consensual sexual touching when “shared to raise awareness (without entertainment or sensational context).” However, “sensational context” in this particular instance is not defined in any way, either internally or in public-facing language, leaving it open to broad interpretation and ultimately to the erroneous removal of content. The lack of clarity may also lead to a failure to enforce content that exploits the awareness-raising content exception for engagement purposes. Given the concerns about the monetization of emotionally engaging content, the Board asks Meta to provide clarity to reviewers on what is sensationalist content in the context of non-consensual sexual touching, to ensure more accurate review of content.

The Board notes with concern the disconnect between Meta’s external policies and internal enforcement guidance relevant to the Brazil case. Meta states that the content was found to be non-violative under Adult Sexual Exploitation’s creepshot policy line because it did not intend to humiliate, mock, or identify the individual captured. The Board agrees with this conclusion. However, as internal guidance definitively states that any content depicting masturbation in public and either mocks, sexualizes, or identifies an individual - including content intending to condemn the behavior - is violative under the creepshots policy line, the only possible outcome for this content would have been a removal under Adult Sexual Exploitation policy. The lack of flexibility in the internal facing policies creates confusion for content reviewers, as it is counter to the rationale in the public facing policy which recognizes the importance of the platform to survivors and others drawing attention to sexual violence and exploitation.

Additionally, the disconnect between the external policies and the internal guidance surfaces another risk. The Adult Sexual Exploitation Community Standard explicitly states that the company allows “survivors to share their experiences,” serving, at least partially, as justification for the awareness-raising policy exception. Users posting content that would fall under the creepshot policy, under the assumption that it fits the awareness-raising exception may then face unintended consequences, such as a standard or severe strike because they are unaware of the internal guidance. These strikes come with additional consequences, such as feature limits, additionally hindering the user’s presence on the platform. The potentiality for users to face disproportionate strikes due to obscure policy guidance from Meta emphasizes the need for a broader and clearer awareness-raising policy exception.

The Board also notes that a warning screen was ultimately applied to the Brazil case content even though doing this was outside the scope of the Adult Nudity and Sexual Activity Community Standard. The restoration of the Brazil case content under the Adult Nudity and Sexual Activity policy required this ad hoc application of the warning screen, though Meta itself stated this was “outside the letter of the policy.” The Board understands that this decision is in accord with the restrictions on awareness-raising content under the Adult Sexual Exploitation Community Standard. However, as Meta correctly chose to consider the Adult Nudity and Sexual Activity Community Standard as applicable to this content, any consequences should be provided for under this policy. While application of a warning screen may be an appropriate policy response to the content, this should be specified in the policy, similar to the way it is specified in the Adult Sexual Exploitation policy.

II. Legitimate Aim

Any restriction on freedom of expression should also pursue one or more of the legitimate aims listed in the ICCPR, which include protecting rights of others (Article 19, para. 3, ICCPR).

Both the Adult Sexual Exploitation policy and the Adult Nudity and Sexual Activity policy aim to prevent abuse, re-victimization, social stigmatization, doxing and other forms of harassment. As emphasized in the India Sexual Harassment decision, they serve the protection of the right to life (Article 6, ICCPR), the right to privacy (Article 17, ICCPR), and the right to physical and mental health (Article 12, ICESCR). The Adult Sexual Exploitation policy also pursues the goal of preventing discrimination and gender-based violence (Article 2, para. 1, ICCPR, Article 1, CEDAW). Aspects of these policies also aim to protect the child from information and material injurious to his or her well-being (Article 17, CRC). These objectives are largely aligned with protecting the rights of others, which is a legitimate aim.

III. Necessity and Proportionality

Under ICCPR Article 19(3), necessity and proportionality requires that restrictions on expression “must be appropriate to achieve their protective function; they must be the least intrusive instrument amongst those which might achieve their protective function; they must be proportionate to the interest to be protected” (General Comment No. 34, para. 34).

In assessing the necessity and proportionality of potential measures, the Board considered the following: (a) the awareness-raising intent references in both posts (either in captions and/or text overlay); (b) the potential vulnerability of the victims in the videos; (c) the vulnerability of the users (specifically, survivors of sexual violence and/or children) who may encounter this material while on-platform.

In these cases, the Board finds that removal would not have been necessary and proportionate in either case because they serve awareness-raising purposes. Both posts did not reveal the identities of the survivors and contained clear signals in the captions that they were raising awareness on sexual harassment and violence.

The Board believes that in both cases the warning screen was a necessary and proportionate restriction on freedom of expression. As stated in the Board’s previous decisions, a “warning screen does not place an undue burden on those who wish to see the content while informing others about the nature of the content and allowing them to decide whether to see it or not” (see Sudan Graphic Video). Warning screens allow users to have critical conversations on gender-based violence and women’s safety while simultaneously alerting other users of the potential sensitivities of the discussion.

However, the Board finds that automatically coupling the “Mark as Sensitive” warning screen with the non-recommendable function is an unnecessary additional restriction on freedom of expression in this instance. Social media plays a significant role in education and advocacy around social issues, including those related to changes to legislation. The Board is concerned that by making this content non-recommendable, Meta is ignoring the value of the platform for survivors and advocates and ultimately undermining the purpose of the awareness-raising policy exception. The Board has previously found that the exclusion from recommendation of awareness-raising content placed behind a warning screen was not a necessary or proportionate restriction on freedom of expression ( Al-Shifa Hospital;Hostages Kidnapped from Israel;Candidate for Mayor Assassinated in Mexico). As further outlined the Eating Disorders decision, Meta’s systems should not impede such content from reaching the intended audiences, especially when the content is visible only to adult users.

The Board agrees with Meta’s decision to enact an age restriction on these pieces of content with the objective of protecting minors from injurious material while allowing awareness-raising content to remain on platform. As previously outlined in the India Sexual Harassment decision, the Board recognizes the importance of protecting children from viewing age-inappropriate and potentially harmful content on the platform. As General Comment No.25 on Children’s rights in relation to the digital environment states: “parties should take all appropriate measures to protect children from risks to their right to life, survival and development. Risks relating to content [...] encompass, among other things, violent and sexual content ...” (para. 14).

In both cases, the Board finds that removal of the content would not be necessary and proportionate, but that applying a warning screen and age restriction satisfies this test.

6. The Oversight Board’s Decision

The Board overturns Meta's decision to take down the content. Both pieces of content should have been left up with the appropriate restrictions of age-gating and warning screens. However, neither piece of content should have been made non-recommendable.

7. Recommendations

A. Content Policy

1. To ensure that content with awareness-raising intent is properly enforced under the Adult Sexual Exploitation policy, Meta should expand the language of the policy exception from “content depicting non-consensual sexual touching” to “content depicting non-consensual sexual touching and/or other forms of sexual violence.”

The Board will consider this recommendation implemented when Meta updates its Adult Sexual Exploitation Community Standard to reflect this more expansive definition of sexual harassment incidences shared for awareness-raising intent.

2. To ensure that content with awareness-raising intent is properly enforced under the Adult Nudity and Sexual Activity policy, Meta should develop a policy exception for awareness-raising, especially for content depicting masturbation in public spaces. This exception should have the same consequences as the Adult Sexual Exploitation policy exception, which include age-gating and warning screens.

The Board will consider this recommendation implemented when the Adult Nudity and Sexual Activity Community Standard is publicly updated to reflect this new exception.

Enforcement

3. To ensure that content that raises awareness is protected, Meta should align its internal guidance and external language on the “creepshots” policy line of the Adult Sexual Exploitation Community Standard to clarify that not all depictions of public masturbation are violative.

The Board will consider this recommendation implemented when Meta shares updated language for the Adult Sexual Exploitation Community Standard internal guidance for “creepshots.”

4. To allow for users to widely share awareness-raising content and still simultaneously allow them to make informed decisions about content they interact with under the Adult Sexual Exploitation and Adult Nudity and Sexual Activity community standards, Meta should decouple the “Mark as Sensitive” warning screen from the non-recommendable enforcement action.

The Board will consider this recommendation implemented when Meta shares with the Board the updated enforcement practices for the "Mark as Sensitive" warning screen when applied to awareness-raising content.

*Procedural Note:

• The Oversight Board’s decisions are made by panels of five Members and approved by a majority vote of the full Board. Board decisions do not necessarily represent the views of all Members.
• Under its Charter, the Oversight Board may review appeals from users whose content Meta removed, appeals from users who reported content that Meta left up, and decisions that Meta refers to it (Charter Article 2, Section 1). The Board has binding authority to uphold or overturn Meta’s content decisions (Charter Article 3, Section 5; Charter Article 4). The Board may issue non-binding recommendations that Meta is required to respond to (Charter Article 3, Section 4; Article 4). Where Meta commits to act on recommendations, the Board monitors their implementation.
• For this case decision, independent research was commissioned on behalf of the Board. The Board was assisted by Duco Advisors, an advisory firm focusing on the intersection of geopolitics, trust and safety and technology.

Return to Case Decisions and Policy Advisory Opinions