Social media platforms have made violence and harassment easy to perpetrate, and difficult to investigate. These platforms provide multi-faceted anonymity for offenders and an environment for these crimes to thrive. While both men and women experience online-based gender violence, women, this investigation and other sources have established, are more likely to be victims of this technology-facilitated crime. 

UN Women defines tech-facilitated gender-based violence as “any act that is committed or amplified using tools or technologies causing physical, sexual, psychological, political, or economic harm to women and girls because of their gender”. Further, it says that these acts are part of a larger pattern including intimate image abuse, doxing, trolling, sharing deep fakes, and misogynistic hate speech among others.  

Further, according to the Council of Europe, misogynistic or sexist hate speech consists of expressions which spread, incite, promote or justify hatred based on sex. 

While there are disparities in statistics on the impact of technology-facilitated violence, it is important to explore how and why this phenomenon thrives. In a study focusing on the top 51 countries by the number of people online, the Economist Intelligence Unit found that by 2020, 38% of women admitted to having a personal experience with online gender-based violence. This report further shows that 65% of women knew someone who had experienced tech-facilitated gender-based violence, and an overall 85% witnessed violence against other women online. 

Source: The Economist Intelligence Unit

A similar report by Plan International focusing on 22 countries shows that in the same year, 58% of young women and girls admitted to having experienced online harassment on various social media platforms. 79% of the respondents thought their harasser was probably male, while 29% thought theirs was probably female. 

According to Data Reportal, at the start of 2023, Kenya’s internet penetration was 32.70% with about 17.86 million internet users. Of these, 55.7% were male, while the remaining 44.3% were female. In 2024, Kenya’s internet penetration stood at 40.8%, with 56.8% male users and 43.2% female. Broken down by platform, in 2024 Facebook had 13.05 million users in Kenya while Twitter, now X had 1.87 million users. The internet has become an integral part of how people communicate, connect and access information in their daily lives. While this is seen as a net positive, there are some negative consequences involved in this shift from the offline to the online space. 

Source: DataReportal

Gender-based violence is a weighty topic due to its pervasive nature, and how it infringes on human rights. As technology and access to internet connections become more ubiquitous, so does tech-facilitated violence, making the fight against online gender-based

violence (OGBV) complex.  Estimates show that one in every three women will experience either physical or sexual violence in their lifetime. Women, according to the World Health Organization, are at a higher risk of gender-based violence. Further, the Gender Violence Recovery Centre (GVRC) estimates that between 39% and 47% of Kenyan women experience GBV in their lifetime.  According to GVRC, this is one of the highest rates globally. 

Tech-facilitated gender-based violence takes different forms, with the most dominant being online sexual harassment, manipulation of information, slut-shaming, and sexist or misogynistic attacks. Technologies such as artificial intelligence have made the problem worse, as is seen with the production of deepfakes and other harmful synthetic media. Technology is facilitating violence in a way that seeks to silence women online, and to extend this violence into the offline space.

Technology has had a significant role in the spread of gendered disinformation, which the US State Department broadly describes as a subset of misogynistic abuse and violence against women that uses false or misleading gender and sex-based narratives, often with some degree of coordination, to deter women from participating in the public sphere.

The Internet Governance Forum further states that this kind of gendered violence targets individuals with higher public status, adding that the primary burden of tackling gendered disinformation needs to fall on the social media platforms amplifying the harms rather than on the women targeted by such campaigns and social media users viewing them.

This investigation focuses on selected Kenyan women in the public eye. It looks to expose the forms that such crimes take, and the role of social media in amplifying these actors. What we found out in the course of this investigation is that online misogyny often manifests as trolling – the deliberate targeted sharing of provocative and offensive posts on the internet, or doxxing which is the unauthorized disclosure of an individual’s information to the public. 

The Bullying and the Trolls

Twitter (now X) alone has had several trending hashtags going up with discussions of different Kenyan women, often. Some of these tweets contained altered images of the First Lady, referring to her overtly Christian public image, but manipulated to exaggerate her features.  

This tweet, for instance, has a picture of President William Ruto appearing to react to the altered image of the First Lady, and the quote, ““Backhots hapo inakaa kuweza”, refers to the sexual nature of the manipulated image. 

Others replied to tweets containing images of the First Lady had the same or similar altered images and made more sexual references. The image manipulated in these tweets was taken during a prayer session for peace in the North Rift region of Kenya, on June 22, 2023. 

Further, the posts containing the altered image were used to reply or quote posts shared by legitimate media houses. or those altered to push indecent narratives, a tactic designed to make the posts more visible.

Any form of gender-based violence often involves silencing the victims, whether through sexualised comments or physical threats. Often, perpetrators target sensitive aspects of a woman’s life such as her body, relationships, marital status, family, sexuality, financial status, and social life. 

Screengrabs of altered images of First Lady Rachel Ruto

Another instance of targeted harassment is that of nominated Senator Gloria Orwoba. On February 14, 2023, Senator Orwoba attended a senate plenary session in ‘period-stained’ pants as part of a campaign to raise awareness of period poverty, which is when women have insufficient access to menstrual products, education, and sanitation facilities. 

Halfway through the session in the Senate, Senator Orwoba was asked to leave the proceedings. According to a point of order raised through the speaker, she was “inappropriately” dressed. Although she was not entirely dismissed from the senate, Senator Orwoba was asked to return only after changing into clothes that were not stained. Meanwhile, some online posts supported the senator for her boldness, while others condemned her move as an embarrassment, and disgusting. Those deriding the senator said that she “behaves like a mad woman”, and others claimed that she was seeking attention.  

Screengrab of comments alongside images of Senator Orwoba

Period shaming is rooted in the belief that publicly discussing periods is taboo, a prevalent issue in many African societies that reinforces gender-based discrimination. Periods are viewed as dirty and disgusting, and this is reflected in the comments and posts targeting Orwoba’s efforts to end period stigma. Related to this, Facebook posts shared by Senator Orwoba received sexualized replies. This Facebook post with the hashtags #EndPeriodShaming and #EndPeriodPoverty shows that that the campaign, which started on Twitter, continued on Facebook. It is in the comments that the bullying thrives

Comments on Orwaba’s posts

Randomly selected posts with the terms ‘Orwoba’, ‘Menstrual,’ ‘Period’ ‘Shame’, and ‘Stigma’ show  the overwhelmingly negative nature of these comments which reinforce the weaponization of shame and silence. 

Comments insulting Senator Gloria Orwoba
More comments insulting the Senator

Another visible target of online misogyny is Azziad Nasenya, a Kenyan content creator, and social media personality, who has frequently shown up on Twitter’s trending section. While this list of trending topics and terms provides a quick glimpse of what people on the platform are talking about, it can also highlight targeted harassment. Due to her celebrity status, Azziad’s life often comes under intense public scrutiny.  She has been the target of frequent tech-facilitated crimes, from impersonation through imposter accounts using her name and photos, to sexualised comments and defamation. Although these comments seem random, they can be classified as cyberbullying. 

Many of Azziad’s social media posts have replies body-shaming her, while others are tweets mentioning and tagging her with comments objectifying her. In one reply, for instance, a user refers to her as, “an international prostitute.”

Posts and replies objectifying Azziad
Replies to Azziad’s tweet

Azziad has official pages on Facebook, Twitter, TikTok and YouTube. However, Piga Firimbi discovered a pool of accounts impersonating her on all these platforms, amplifying inaccurate or malicious information in the process. Some of these accounts were identified as parody or impersonator accounts, using Azziad’s image to gain free publicity.   

In her Engage Talk, Azziad opened up about how badly she was affected the first time her contact information was shared publicly, just as she was gaining fame from a TikTok video she had made. 

“It took me 19 years to become an internet sensation. So one day, my number leaks”, she narrates. “The calls started coming in. I had to bar all calls. Then the messages start streaming in. Some of them were encouraging which I am grateful for. Some of them were flirtatious which I ignored. The ones that caught my attention,” she says amid tears, “were the cruel and unkind ones.” 

Azziad is very familiar with how cyberbullying manifests. 

“You know, I always read about cyberbullying in school. I never thought about it, I never knew it was something that would happen to me or anyone around me.”

Cyberbullying and sharing one’s opinions manifest in similar ways. What is considered trolling by a section of internet users can be regarded as online humour by another, often referred to as ‘banter,’ in casual terms. 

On the other hand, memes have contributed immensely to this form of gender-based violence. Exploring the impact of online gender-based violence through a victim’s experience shows that there is a difference between sarcasm from cyberbullying. The use of memes for sarcasm is common, but it turns into cyberbullying when it intends to perpetuate a toxic online environment. 

These posts seem to suggest that Azziad randomly has affairs with politician, often attaching images from movie scenes with sexual captions. This one for instance reading, “Azziad to every politician,” includes an image from a website with explicit adult content. Another with the same caption attaches a screengrab with similar descriptions. 

Comments on Azziad’s post with pornographic pictures

The Computer Misuse and Cybercrimes Act, of 2018  does not explicitly define cyberbullying as a standalone offence. However, Section 27 of the Act addresses cyber-harassment.  It states that a person commits an offense if they knowingly communicate in a way that causes detrimental harm, fear, or offense to another person or their property. 

Section 37 of this Act adds that if a person wrongfully distributes obscene or intimate images, “including making a digital depiction available for distribution or downloading through a telecommunications network or through any other means of transferring data to a computer”, the intimate or obscene image of another person commits an offence and is liable, on conviction to a fine not exceeding two hundred thousand shillings or imprisonment for a term not exceeding two years, or to both.” 

In an interview with Piga Firimbi, Azziad admitted that she finds fighting bullies irrelevant, especially when you do not have a face you are responding to. She adds that, in instances where she has had to fight back, she has fought back with the law.

This tweet shares an altered image of President William Ruto shaking hands with Azziad Nasenya with more attention on his shoes. The post reads, “Aziad is trending because of shoes”, even though she had been pictured wearing the shoes before

Altered image of Azziad and Ruto

Twitter’s Abuse and Harassment policy states that the company is committed to safe online discourse. The policy states in part, “On X, you should feel safe expressing your unique point of view. We believe in freedom of expression and open dialogue, and to facilitate healthy dialogue on the platform and empower individuals to express diverse opinions and beliefs, we prohibit behaviour and content that harasses, shames or degrades others,” 

Violations of this policy as identified by the platform include the use of repetitive or isolated insults or profanity with the intent to harass or intimidate someone, soliciting sexual acts, unwanted sexual discussion of someone’s body, and continuously sharing posts with malicious content, dedicated to harassing an individual.

Twitter does allow for some consensual nudity and adult content, provided that this content is marked sensitive. 

However, we identified profanity and vulgar language of a sexual nature that was prevalent on the platform. The terms used were mostly a combination of English-Swahili slang (Sheng), which makes it difficult for automated systems to identify and flag them as being offensive. 

Graphic showing demeaning words and phrases targeting Azziad

Using Gephi, we found 11,227 tweets, retweets, and quoted tweets identified as either misogynistic or defamatory, and that have interacted with posts labeled as online gender-based violence. These interactions include publishing, commenting on, retweeting, or liking such posts. The larger names show those which received the most retweets, indicating a wider reach and influence of these particular messages within this network.

Network diagram of accounts posting and interacting with online gender-based violence content

This post must have appeared on your timeline between September 17 and September 18, 2023. It is a tweet by Hannah Gitau, alias Annitah Akiru Raey, a media personality, feminist and formerly a radio presenter at Ghetto Radio, Hot 96, Radio Jambo, and Milele FM.

Annitah Raey’s original tweet

Across most of her social media pages, Annitah shared that she was diagnosed with Bell’s palsy, a medical condition that causes paralysis on one side of the face. She had been dealing with it for the better part of 2023. On TikTok, where she first posted about her diagnosis, she wrote, “fell sick a month ago and it’s been a roller-coaster. I am getting better and healing from Bell’s paralysis. I am grateful for life honestly. For family and for everyone who has held my hand through this.”

The replies to this post on TikTok were largely positive. About a month later, she shared an update on her progress on Twitter. However, unlike with her first TikTok video, this update was received with hostility. More than half of the quote retweets and replies to this post were very provocative. 

For instance, this quoted reply says, “I hope it gets worse. Always remember those days when you thought you were unstoppable. You called us dogs… Now who’s the dog?”

Quote reply to Annitah’s tweet with insulting language

The trolling spread to some of her previous posts, and it is clear that those behind the posts did not care about how much harm their posts could cause. 

One such reply read, “Even a broke man will not come close to a person with her mouth close to her cheek.”

A similar quote retweet went as far as suggesting possible death from her condition. It read, “I sell coffins, DM for business.”

Archived links here, here, here, here and here show more of such quoted replies. 

Screengrab of replies directed at Annitah
More replies directed to Annitah’s original tweet

As expected, these and other replies to Annitah’s post were reposted on Facebook, thus, amplifying the damage on Twitter. One post, for instance, shared four different screenshots from Twitter. Out of the 26 comments to this Facebook post, one addressed these screenshots which later elicited a back-and-forth in the comment section. Using the same screenshots, two other posts still on Facebook, advised, “Stop sharing your problems on the internet, media personality Anita Raey is being nuked huko Twitter.” 

Tweets and Facebook posts targeting Annitah

Posts here, here, here, and here are examples of deriding comments targeting Annitah. 

This is not Annitah’s first experience, and she recalls her first encounter with trolls in 2018, when she was a radio host at Ghetto Radio. 

In a phone interview with Piga Firimbi, she says that those trolling her were mainly women, and the comments she received were mainly focused on things that she held to be very personal – her body, sexuality and marital status. Because of this, Annitah had to take a few days off work to seek professional help. 

Much as these comments came from faceless and anonymous pseudo accounts, what got to Annitah were the comments made about her personal life containing details she says she has intentionally kept off social media. 

Her biggest concern is the anonymity that X offers, and the nameless, faceless accounts attacking her. 

“It was not fair because I get to be on the public stand, because I am a media personality, but you get to hide!”, she laments. 

The trolling on Twitter worsened when anyone could get verification. Until November 2022, you needed to earn the prized blue verification check mark by providing evidence of some degree of prominence and authenticity. The platform’s new owner, Elon Musk, changed these requirements, needing only a verified mobile number and a paid subscription to the Blue Plan for one to be verified. 

According to CNN’s tech reporter Brian Fung, Elon Musk’s decision to roll out a paid verification option as part of Twitter’s subscription product, and removing legacy blue checks from accounts, has led to unintended consequences, including a wave of troubling impersonations and the potential for new scams and misinformation.

Fung also adds that verification is an even less transparent and meaningful indicator of a person’s identity.

Annitah believes that if Twitter wanted, it would devote resources towards limiting trolls on the platform. The fact that Twitter allows pseudo-accounts, she states, is the biggest challenge for users of the platform in her opinion.

Her tweet sharing the Bell’s palsy diagnosis had almost 1.7 million views. None of the comments, replies and quoted replies bothered her. However, she noticed that the online posts from people that stepped in to defend her ended up receiving the trolling to themselves, and this got her attention. 

“Trolling is something else. Sometimes it is worse for the people around you than it is for you”, she says, adding, “If my boldness intimidates you, that is a you problem, and I refuse to engage with trolls.”

Research by Trust Lab in in three EU member states – Spain, Slovakia and Poland, shows that Twitter has the largest portion of mis/disinformation posts and equally, the highest number of disinformation actors.

Initially, the participants in the pilot study of this report were Facebook, Instagram, LinkedIn, TikTok, Twitter and YouTube. These platforms were voluntarily signatories to the EU’s Code of Practice on Disinformation. Halfway through the study, Twitter pulled out from this code of practice. Despite this withdrawal, the platform is still subject to the EU’s Digital Services Act (DSA).

The report admits that online disinformation is an ambiguous and fast-changing phenomenon, and measuring disinformation is challenging.

There is no doubt that Online Gender-Based Violence is built on aggressive comments, unwanted sexual remarks, and gendered disinformation which is slowly evolving into more complex forms, as a consequence of artificially generated content. 

Angela Minayo, an advocate of the High Court and the Lead Programs Officer at Kenya ICT Action Network (KICTANET) believes that since there is a culture of normalising abuse towards women and girls when they experience online violence, society often first questions what they did to warrant this kind of violence. 

Women, Angela says, are sometimes attacked just for existing online, sending the message that they should not be in these spaces.

Gendered violence, online or offline cannot exist without attitudes and beliefs which minimise and normalise these kinds of crimes. Simply tolerating these biased attitudes, and microaggressions eventually excuses sexual violence.

Ultimately, while violence against women and girls is one of the most prevalent forms of human rights violations, a report by the Social Development Direct regards tech-facilitated violence as the least reported form of GBV, and when reported, it is retracted or amicably resolved. Sexual victimization is a barrier to having these cases reported. 

There is yet another challenge to prosecuting this kind of crime, as Angela reveals. 

“Even as a lawyer,” she says, “prosecuting tech-facilitated violence is very hard. There is a lot of context needed by the court to understand the gravity of what has happened. The reporting is still low, and we don’t have enough jurisprudence.”

She adds that while there is a framework for implementing the Computer Misuse and Cybercrimes Act of 2018, there needs to be more training on these laws, and prosecuting these crimes in court or even handling such clients. 

Ultimately, identifying toxic content is not a straightforward task, a UNESCO report on Technology-Facilitated Gender-Based Violence says. The report calls for a multistakeholder approach when developing, disseminating, and generating policy, education, infrastructure, and technological approaches to address the potential and already realised harms of generative AI‑enabled technology facilitated gender‑based violence.

The Commission on the Status of Women (CSW) recommends strides towards eliminating, preventing and responding to all forms of violence against women and girls, online and offline. It suggests using multisectoral and coordinated approaches to investigate, prosecute and punish the perpetrators of violence and impunity.  

These approaches should be undertaken to to create safe, and violence-free working environment for women, including by ratifying key international treaties that protect against gender-based violence and sexual harassment. This can be achieved by condemning and taking all appropriate measures, including legal action, to combat the use of digital tools, social media, and online platforms, to produce and publish any form of tech-facilitated violence against women and girls. 

As governments strive for a safer digital landscape, the Commission suggests that governments should, promote and invest in gender-responsive digital and science & technology education in the digital age and encourage efforts towards putting children’s needs, in particular those of girls, at the centre of digital policy, where relevant.

The Commission recommends investing in teaching digital and data literacy and proposes a digital literacy curriculum that would integrate with the national curricula at all levels to teach digital and data literacy. These efforts aim to equip all women and girls with transferable skills to ensure they are both safe and empowered in using digital technology for leisure, education, information, and reporting all forms of violence. This includes, teaching on competency and skills related to digital technologies in both formal and informal educational settings, to raise awareness and foster understanding on ethical and responsible online conduct. 

As emphasized in a report by Plan International, online gender-based violence is a human rights issue, it also highlights that girls and young women are sick of being harassed, in some cases being pushed away from all the opportunities that the internet could provide, in a better world. As all the cases of online gender-based violence have shown, the online spaces that were meant to bring us together have been used to direct violence against women online, and addressing this will need the intervention of tech platforms, governments and regulators to address. While much has been achieved in getting more people online, a lot more needs to be done to ensure that the internet is safe for everyone. 

Are you the victim of harassment and gender-based violence online? Here’s a list of organisations that can help.

If you are going through abuse or you know a person going through abuse, the free Gender-Based Violence hotline is 1195. 

This article was produced with mentorship from the African Academy for Open Source Investigations (AAOSI), to tackle disinformation that undermines our democracies, as part of an initiative by the International Centre for Journalists (ICFJ) and Code for Africa (CfA). Visit disinfo.africa for more information.

This article was edited by Piga Firimbi Fact-checking Editor Calvin Rock

Add comment

Your email address will not be published. Required fields are marked *