Tech Harms During Elections in Africa

May 2024

By Bulanda T. Nkhowani – Campaigns and Partnerships Manager for Africa- Digital Action.

Table of Contents

1. Introduction
2. How online harms manifest during election cycles
Section 1. Misinformation and Disinformation
Section 2. Undermining confidence in democratic processes and targeting of electoral officials or prominent voices
Section 3. Hate speech and election violence
Section 4. Meddling by foreign actors and international influences
Section 5. Gender-based violence and harms against minorities
3. Recommendations and Conclusion


The year 2024 is dubbed the most significant year for democracy[1] in this century with over 65 countries around the globe scheduled to go to the polls. In the African region, over 18 countries will vote, including the biggest democracies like South Africa, Senegal and Ghana, and fragile democracies like Mozambique and South Sudan.

The African region has undergone rapid digital uptake and transformation coupled with the emergence of cybercrime, cybersecurity and data protection laws and in some instances introduced frameworks for Artificial Intelligence aimed at safeguarding the online space for users. While technology contributes to tackling numerous social, economic, and political challenges, there has also been growing concern about the accountability of tech companies and the impact of their products and services on society.

In particular, these concerns centre around the impact of technology on democracy. While the internet has boosted civic participation by making leaders and governments more accessible to electorates, enabling citizens to speak truth to power in fast, cost-effective and sometimes anonymous ways, technology has turned out to be a double-edged sword that has presented users with both positive and negative consequences.

In the context of elections, increasingly technology is playing a crucial role in enabling information access and dissemination among voters, electoral officials, political parties and the media, including supplementing political campaigns and enabling an alternative space for expression and contestation. While the battle for democracy may have shifted to the online spaces it is not without any consequences, there is a rise in incidents of mis/disinformation, hate speech, tech-assisted violence against women, privacy violations, trolling and incitement to manipulation of democratic processes and other forms of harmful content that have proliferated the internet and social media platforms during electoral cycles.

Largely, the majority of elections and power transitions on the continent in recent times have been peaceful and encouraging in countries[2] like Lesotho, Kenya, Nigeria and Zimbabwe, however, 2023 unleashed a new wave of military coups d’états[3] that hit the region affecting, Mali, Guinea, Burkina Faso, Niger and Gabon.

Thus, it is crucial to understand the role that technology plays within the electoral cycle and the role that users, platforms, developers and other stakeholders can play in safeguarding the information space during elections.  There is a startling inequity that exists between regions when it comes to protection from harm by tech platforms. Tech companies headquartered in the global north – and the regulators responsible for them – are negligent when it comes to dealing with the impact of technology on the global majority countries.

This paper seeks to highlight the prevalent tech harms witnessed during elections in Africa, how they manifest from country to country and proposes solutions and recommendations that platforms can take to protect electoral integrity and social media users. The paper draws on a targeted survey and expert interviews, the insights will serve as a basis for engaging policymakers and big tech companies as part of a broader campaign on tech accountability.

How online harms manifest during election cycles

Social media platforms Facebook, Twitter, WhatsApp, YouTube and TikTok are widely used in the region. The survey singled out Facebook, Twitter and WhatsApp as being the most prone to dis/misinformation campaigns during elections in the region. On the other hand, digital rights activists, election workers and journalists identified threats such as incitement, hate speech, misinformation, meddling by foreign actors and international influences, tech-facilitated election violence, online gender-based violence and harms against minorities as the most prominent threats affecting the region. In addition, undermining confidence in the electoral process and integrity, and targeting of electoral officials, political figures and prominent voices.

It is also important to note that offline threats to electoral integrity are often reinforced online using electronic means. Traditionally, elections were characterised by persuasive offline propaganda campaigns that have now moved to the online space in the form of disinformation, meddling by foreign actors and undermining of trust in electoral officials and processes.

Misinformation and disinformation

Misinformation and disinformation are a great threat to democracy and pose challenges to the authenticity of information circulated during election cycles. Disinformation campaigns seeking to manipulate African information systems have surged[4] fourfold since 2022.  Due to mainstream media’s limited coverage of the elections and the time taken to verify and fact-check information, for instance, in Kenya, people mainly rely on social media for timely updates about the elections. Dis/misinformation is not new to the Kenyan electoral information landscape. From as far back as 2017, Kenya was plagued by reports of fake news during elections, however, the 2022 polls saw the circulation of false information that played on existing beliefs, fears and biases.

In the run-up to the 2022 Kenyan elections, political disinformation was rife both offline and online amid competition for popularity and non-public interest agendas and took the form of texts, images, video or audio content. During the campaign period[5], fake news and videos targeting less informed citizens were disseminated over TikTok X (ex Twitter) and WhatsApp, making distinguishing authentic news from falsehoods very difficult. Influencers (bloggers, journalists and other digital experts, communications specialists etc.) were hired by political parties to shape public opinions online about their political manifestos and false narratives depicting their preferred candidate as leading in the polls.

Fraudulent opinion polls attributed to legitimate sources such as GeoPoll and The Daily Nation newspaper became a major trend. Graphic footage and images[6] depicting the police acting brutally toward protestors were circulated over social media during the election period seeking to fuel tension in the country. Websites mimicking genuine news outlets disseminated fake stories and propaganda about candidates. The use of coded language like the Swahili word “madoadoa”[7] (meaning “blemish”) posed major challenges for social media platforms to moderate content effectively as they lacked an adequate understanding of the context in which such words were used. An investigation by Mozilla uncovered weaknesses in TikTok’s[8] content moderation policies coupled with content moderators’ lack of capacity to understand the Kenyan political content that allowed for disinformation and violent ethnic narratives to spread across the country before, during and after the elections. For instance, a video was circulated on TikTok that contained a manipulated image of one of the political candidates wearing a shirt that was covered in blood, holding a knife to his own neck with a caption alleging that he was a murderer. The video received over 500,000. Prior to voting false claims of military deployment in the capital Nairobi, wild animals on the loose in certain regions and claims of candidates that had allegedly already won the election circulated on social media.

Similarly, in the Nigeria 2023 election, social media influencers were secretly hired[9] and paid by political parties to peddle misinformation online about political opponents. The influencers were given cash, lavish gifts, government contracts and even offered political appointments for their work. Misinformation and disinformation proliferated social media platforms[10] throughout the electoral cycle, false written posts, photos and videos about political candidates were shared both openly and privately on platforms by political parties and their supporters.

In Zimbabwe, WhatsApp was the channel of choice[11] for disinformation campaigns because it provides a broader reach and many Zimbabweans consider it a safe means of communication where they can criticize the government and avert the risk of being arrested. Twitter was awash with bots and accounts with pseudonyms that were commissioned to push the government’s accomplishments. In the Democratic Republic of Congo, disinformation cases[12] in the 2023 elections were mainly targeted at discrediting the electoral process and opposition candidates.

Interestingly, the Ugandan government banned access to Facebook[13] ahead of the 2021 elections after the company closed some pro-government accounts that were spreading political disinformation. Facebook remains banned in the country, only accessible using VPNs and talks between the government and Meta to lift the ban remain stalled.[14]

Undermining confidence in democratic processes and targeting of electoral officials or prominent voices

In the aftermath of the 2023 Nigerian election opposition political parties petitioned the election results. Social media carried an explosion of fake news, [15] some of which targeted the electoral commission and Supreme Court judges. The fake news included false claims that a minister of the winning party had access to the computer systems containing election results prior to the winner being announced. In another incident, a false image was shared depicting the President-elect Bola Tinubu bribing the Chief Justice of the Supreme Court which in actual fact was an old image of the two together in 1996. Another video depicted the president of the electoral commission dining at the homes of the President-elect a day after the elections, this video was confirmed to be false by local fact-checkers.

In Zimbabwe, the 2023 election saw disinformation campaigns attempting to discredit the electoral commission[16] using false reports that the polling agents and election observers were being recalled to sign the new V.11 forms (forms used by electoral management bodies to record and process election results), thereby alleging that the original V.11 forms had been replaced. Foreign election observer missions[17] like the European Union observer mission, the Carter Centre, and the Southern African Development Community (SADC) were equally victims of the disinformation smear campaigns.

Hate speech and election violence

When hate speech and disinformation spill off from online platforms into the offline world, they contribute to real-world harms. During the 2023 DRC elections, prominent well-known individuals contributed to the spreading of messages containing ethnic hatred and violence on social networks. A report by Internet Sans Haine recorded over 1188 messages containing hate speech, violence and disinformation [18]from monitoring activities conducted on Facebook, Twitter, TikTok and YouTube and observed from October 2023 to January 2024. Consequently, about  19 deaths[19] were recorded as a result of electoral violence of which online hate speech was identified as a key contributor. A report[20]found that African countries experiencing active conflict are more prone to greater levels of disinformation that lead to real-life violence thus highlighting the connection between instability and disinformation.

Similarly, in the Lagos elections, anti-Igbo rhetoric laced with ethnic undertones and stereotypes were targeted at the non-Yoruba-speaking communities, calling them visitors and landgrabbers. Top government officials,[21] also made similar posts on social media and through lengthy blogposts including taking part in an ethically divisive protest,[22]also alleged that ‘certain people’ were trying to take over Lagos state from the perceived original inhabitants and they could stop this by voting in the APC Party. A study conducted during the Uganda elections indicated that the internet shutdown[23]experienced during the 2021 elections hindered stakeholders and opposition supporters toeffectively challenge electoral malpractice.[24] Platforms like Facebook remain blocked[25] in Uganda and can only be accessed using virtual private networks (VPNs).

Stoking of hate speech on online platforms is not new in Africa. During the civil war in the Tigray region of Ethiopia between 2020 and 2022, Meta was accused of being unresponsive[26] to removing hateful and inciting speech that fuelled deadly ethnic killings in the region. Meta was cited as having fuelled conflict in the region where thousands died and millions have been displaced since fighting broke out between government forces and armed opposition groups. Some fact checkers, civil society and human rights organisations confirmed that their requests for assistance were ignored and any attempts to meet the platform representatives failed to materialise. In addition, thelocal groups[27] doubted if the company had invested enough in content moderation efforts in Africa. Hate speech coupled by disinformation filled posts played a key role in the killings of people[28] including the killing of Tigrayan University chemistry Professor Meareg Amare. Some posts contained names of individuals and photos that helped create an atmosphere “inciting attacks, killings and displacements”. By calling for a “cleansing” of the Amhara territories of the Targaryen. Whistle-blower Frances Haugen had earlier warned that Meta was fanning violence[29] in Ethiopia. Civil society groups have sued Meta[30] for fuelling ethnic violence in Ethiopia and demanding that the company stop Facebook’s algorithms from recommending violent and harmful content to Facebook users and compel Meta to create a Ksh 200 billion ($1.6 billion USD) victims’ fund.

While the 2022 Kenyan election was marred by an onslaught of hate speech and disinformation online, these did not escalate into to violence[31] like was witnessed in 2007 – 2008 when post-election violence broke out due to false information online. This development is attributed to the multi-sectoral approach, where fact-checkers, content moderators, civil society organisations, governments, media stations and citizens all worked to curb hate speech and disinformation.

Meddling by foreign actors and international influences

An investigation[32] found that in the run-up to the 2022 Kenyan elections, an Israeli disinformation specialist hacked into email and social media accounts of political advisors close to President William Ruto. While the hacking did not have a direct influence on the poll or President Rutos victory, it highlights the growing action of the involvement of bad or foreign actors and paid operatives that pose risks to the sovereignty of elections, democratic systems and institutions across Africa.

Another study[33] found that a British data analytics firm is suspected to have used the large-scale data gathered from Kenya’s publicly available voter registration databases and the data it collected from Facebook to conduct online political Microtargeting on digital platforms to sway voters’ decisions. With the requisite data, advertising options on a platform like Facebook can be used to micro-target voters during elections. Some of the ways in which audiences may be affected by microtargeting include invasion of privacy, data breaches, manipulation of voters and voter exclusion. The Africa Centre for Strategic Studies in a report that maps regional disinformation[34] revealed that about 60 percent of disinformation campaigns on the African continent are foreign state-sponsored—with Russia, China, the United Arab Emirates (UAE), Saudi Arabia, and Qatar.

Gender-based violence and harms against minorities

During the January 2021 elections, research that monitored the social media accounts of key women politicians in Uganda[35] concluded that female politicians were more likely to experience targeted and repeated attacks of online violence on Twitter as compared to Facebook and the candidates were likely to experience trolling, sexual violence, body shaming, insults and gendered abuse. Often negative comments[36] on the women candidate’s pages pointed to their inability to hold positions and reminded them of their place outside of politics thus reinforcing many existing stereotypical narratives.

Another report by Pollicy indicates that organized trolling[37] has been on the rise, especially against women with public-facing careers such as journalists, media personalities, activists and politicians.

A prominent activist and co-leader of the “#BringBackOurGirls” movement in Nigeria faced online attacks which intensified during the 2019 Nigerian elections.[38]

Similarly, in Kenya and Zimbabwe, women’s rights groups[39] and the African Commission on Human and Peoples Rights[40] warned of violence against women ahead of the elections. In Kenya after the 2022 elections, a female political candidate vowed to never vie for political office[41] again due to the violence she faced on Facebook while the Kenyan vice-presidential candidate Martha Karua[42] was the target of online attacks based on being a single and unmarried.

In Zimbabwe, a study[43] found that all women leaders interviewed ahead of the 2023 elections in Zimbabwe faced some kind of online violence and higher profiled leaders were more likely to face gendered attacks online leading to negative impacts on their participation and success in politics as some delete their accounts and shun social media and politics entirely.

Gendered disinformation is being used across Africa as a tool to silence critics and exclude women from online civic discourses, especially female politicians and activists and presents reverberating effects on other women and girls who self-censor. Similarly, gendered disinformation was weaponised against the daughter of a Kenyan presidential hopeful, Winnie Odinga,[44] with accusations that she was gay. Disinformation that alleges same-sex relations or that one is gay is highly dangerous in Africa where same-sex relations are criminalised[45] in over 31 countries, sometimes attracting the death penalty.

Recommendations and Conclusion

The numerous elections on the horizon in the African region necessitate the need to protect elections, people and democratic institutions. Our recommendations aim to empower tech platforms and other stakeholders to play a crucial role in supporting election integrity and protecting users in light of the identified harms highlighted in this paper. These recommendations seek to empower platforms to play a proactive and solutions-oriented role in safeguarding electoral integrity and protecting voters.

Drawing upon the Year of Democracy campaign asks,[46] this paper makes a series of recommendations to social media as follows;

  • Provide fully and equitably resourced election plans – that are proportionate to the risk of harm presented by each country and conduct systematic risk assessments of each election at least 6 months before the elections and use these findings to inform election plans. This includes liaising with EMBs, civil society and media in advance of election periods, to inform platforms’ policies, response mechanisms and mitigation strategies against election disinformation.
  • Ensure trustworthiness and information plurality during elections- by establishing working relationships with local trusted partner networks, election management bodies, civil society, fact-checkers etc. who can weigh in with and provide critical information on the political, social, and economic contents of the country. As well as proactively ensure that recommender algorithms do not impede access to accurate electoral information or to diverse points of view within the democratic spectrum but instead coordinate, promote and amplify authoritative electoral information from independent EMBs and public interest and journalistic content.
  • Counter the reach and effects of dis- and misinformation and online gender based violence– by engaging in effective and timely content moderation, determining the number of human content moderators to support the automated content moderation process as well as the language coverage of the content moderators to ensure cultural competency. Platforms should also enhance cross-platform collaboration to prevent the sharing of electoral disinformation across different social networks. In addition, strengthen the accountability of highly influential actors- Put special accountability regimes in place for journalists, influencers and highly influential actors in the election process, along with appropriate sanctions based on published information, and impartial rules for VIP accounts.
  • Establish proper oversight and transparency- platforms should publicly Publish terms of service and policies, and provide regular updates on changes in election-related policies, recommender operations, content moderation systems and modalities for cooperation with trusted flaggers, including information about government requests to take down accounts or requests for user information. Platforms should ensure transparency in political advertising by clearly labelling such content on the platforms and giving users power to opt out of such content. In addition, publishing financial information of platform political advertisers and ad libraries to enable scrutiny of campaign finance and spending.
  • Make data available for researchers- to allow independent researchers and civil society to conduct research to understand effects of social media and online harms on people and elections. As well as eliminate all barriers to accessing platform data to allow for scrutiny by independent actors who are key in framing understanding and informing platform policies and protocols on combating online violations. Social media platforms should dedicate financial and in-kind resources to scale up civil society and researcher’s capacity to document, report on, and analyse online and offline harms in relation to elections.
  • Combat deceptive use of AI- (deepfake and Gen AI content) – while there have been few instances of deceptive uses of AI in the African elections so far, experts warn that the creation, use and distribution of such content will intensify as more people acquire the requisite skills to do so. Platforms must actively work to identify potential risks posed by their recommender systems and use cases during the elections as well as set up concrete guidelines on the use of generative content. Platforms should also work to flag and label gen AI content, especially in Africa, where there is already a high degree of social and economic inequality, algorithmic bias can exacerbate existing disparities and undermine social cohesion.

In closing, the integration of technology into the electoral landscape has brought both benefits and challenges. Elections in the African region have historically been marked by uncertainty and instability, and social media has only exacerbated these dynamics triggering destabilizing and antidemocratic consequences. Social media has become impactful as users are increasingly turning to online platforms as their main source of information.[47] Thus, making the need to understand how social media and the internet influence and impact elections and democracy more urgent. Platforms must play their part to understand their role in shaping society and the potential implications this could have on the democratic fabric and the global community.

Contact: If you’d like to find out more get in touch at: [email protected]


[1] Digital Action (2023) Year of Democracy Campaign website, [Accessed 22 Nov. 2023].

[2] Institute of Strategic Studies (2023) Can the trend of peaceful African elections be sustained?,  [Accessed 22 Nov. 2023].

[3] Al Jazeera (2023) Africa’s coup epidemic: Has democracy failed the continent?,  [Accessed 22 Nov. 2023].

[4] Africa Centre for Strategic Studies (2024) Mapping a Surge of Disinformation in Africa,[Accessed 14 March 2024].

[5] AFP Factcheck, (2022) Election campaigning ends in Kenya but disinformation battle drags on, [Accessed 23 Nov 2023].

[6] Al Jazeera (2022) How Kenya became the latest victim of ‘fake news’, [Accessed 23 Nov 2023].

[7] The Star (2022) ‘Blemish or spot’: Madoadoa phrase explained, [Accessed 23 Nov 2023].

[8] Mozilla (2022) New Research: Disinformation on TikTok Gaslights Political Tensions Ahead of Kenya’s 2022 Elections,[Accessed 23 Nov 2023].

[9] BBC, Nigeria elections 2023: How influencers are secretly paid by political parties, [Accessed 23 Nov 2023].

[10] Public Media Alliance (2023) Nigeria Election: Threats, sanctions, and disinformation, [Accessed 23 Nov 2023].

[11] France 24 (2023) Zimbabwe election disinformation spreads on WhatsApp, [Accessed 23 Nov 2023].

[12] Africa Sans Haine (2024) Report on hate speech related to the 20 December 2023 elections in the DRC, [Accessed 12 Feb 2024].

[13] Anadolu Ajansı (2021) Uganda’s president bans Facebook for being ‘arrogant’,

[14] Monitor (2022), Facebook to remain shut as govt talks with tech giant stall, [Accessed 23 Nov 2023].

[15] Africa News (2023) Nigeria: disinformation reflects a demonetization of institutions, [Accessed 24 Nov 2023].

[16] The Herald (2023) ZEC Dismisses V11 Forms Allegations, [Accessed 24 Nov 2023].

[17] Social Science Research Council (2024) Electoral Disinformation during the Disputed 2023 Elections in Zimbabwe, [Accessed 22 Feb 2024].

[18] Africa Sans Haine (2024) Report on hate speech related to the 20 December 2023 elections in the DRC,  [Accessed 12 Feb 2024].

[19] Actualite CD (2023) DRC: At least 19 deaths related to electoral violence, according to the Carter Center, [Accessed 12 Feb 2024].

[20] Ibid

[21] Vanguard (2023), Lagos, my Lagos!, [Accessed 12 Feb 2024].

[22] Hum Angle (2023) #NigeriaDecides: Hate Speech Fuels Lagos Elections[Accessed 12 Feb 2024].

[23] Associated Press (2021) Internet ‘blackout’ in Uganda on eve of tense election [Accessed 12 Feb 2024].

[24] Journal of Peace Research (2023) Pulling through elections by pulling the plug: Internet disruptions and electoral violence in Uganda [Accessed 12 Feb 2024].

[25] Ibid

[26] Amnesty International (2023) Ethiopia: ‘A death sentence for my father’: Meta’s contribution to human rights abuses in northern Ethiopia, [Accessed 17 Jan 2024].

[27] Amnesty International (2023) Ethiopia: Meta’s failures contributed to abuses against Tigrayan community during conflict in northern Ethiopia, [Accessed 17 Jan 2024].

[28] Ibid

[29] The Guardian (20210 Facebook’s role in Myanmar and Ethiopia under new scrutiny [Accessed 17 Jan 2024].

[30] Amnesty International (2022) Kenya: Meta sued for 1.6 billion USD for fueling Ethiopia ethnic violence, [Accessed 17 Jan 2024].

[31] The London Schoolof Economics and Political Science (2023) Disinformation was rife in Kenya’s 2022 election, [Accessed 17 Jan 2024].

[32]  The Guardian (2023) Political aides hacked by ‘Team Jorge’ in run-up to Kenyan election, [Accessed 17 Jan 2024].

[33] Journal of Intellectual Property and Information Technology Law (JIPIT) (2023) Beyond the Ballot:  A Comparative Analysis of Political Microtargeting Practices and Regulations in Kenya and Nigeria [Accessed 17 Jan 2024].

[34] Ibid

[35] Pollicy Amplified Abuse: Understanding Violence against Women in Politics and Leadership

A study on the 2021 Uganda General Elections, [Accessed 31 Jan 2024].

[36] CIPESA (2022) Digital Threats to Women in Politics : During the Uganda 2021 Election

[37] Pollicy (2020) Alternate Realities, Alternate Internets: Feminist Research for a Feminist Internet [Accessed 31 Jan 2024].

[38] Web Foundation (2022) Online Gender-Based Violence Story – Aisha, Nigeria [Accessed 31 Jan 2024].

[39] Africa News (2022) Kenya: Right groups warn of violence against women during elections, [Accessed 31 Jan 2024].

[40] The Zimbabwean (2023) Women facing ‘offline and online’ violence ahead of elections in Zimbabwe – African commission, [Accessed 31 Jan 2024].

[41] UN Women Africa (2023) Online harassment risks pushing Kenyan women out of politics, [Accessed 31 Jan 2024].

[42] Institute of Development Studies (2024) Weaponising gendered disinformation across Africa [Accessed 15 May 2024].

[43] Advancing Learning and Innovation on Gender Norms (2023) Women in politics in Zimbabwe: how gender norms are fueling online violence, [Accessed 15 May 2024].

[44] Ibid

[45] Amnesty Intonationonal (2024) Africa: Barrage of discriminatory laws stoking hate against LGBTI persons, [Accessed 15 May 2024].

[46] Digital Action (2023) Year of Democracy Campaign Asks,   [Accessed 22 Nov. 2023].

[47] International Institute for Democracy and Electoral Assistance (2021) THE GLOBAL STATE OF DEMOCRACY 2021: Building Resilience in a Pandemic Era,[Accessed 31 Jan 2024].

Share this post