Hostname: page-component-669899f699-ggqkh Total loading time: 0 Render date: 2025-04-25T16:25:14.366Z Has data issue: false hasContentIssue false

Reinstating trust in elections in the era of artificial intelligence and emerging technologies

Published online by Cambridge University Press:  21 April 2025

Fola Adeleke*
Affiliation:
School of Law, University of Witwatersrand, Johannesburg, South Africa

Abstract

Africa had a busy election calendar in 2024, with at least 19 countries holding presidential or general elections. In a continent with a large youth population, a common theme across these countries is a desire for citizens to have their voices heard, and a busy election year offers an opportunity for the continent to redeem its democratic credentials and demonstrate its leaning towards strengthening free and fair elections and a more responsive and democratic governance. Given the central role that governance plays in security in Africa, the stakes from many of these elections are high, not only to achieve a democratically elected government but also to achieve stability and development. Since governance norms, insecurity, and economic buoyancy are rarely contained by borders, the conduct and outcomes from each of these elections will also have implications for neighbouring countries and the continent overall. This article considers how the results of recent elections across Africa have been challenged in courts based on mistrust in the use of technology platforms, how the deployment of emerging technology, including AI, is casting a shadow on the integrity of elections in Africa, and the policy options to address these emerging trends with a particular focus on governance of AI technologies through a human rights-based approach and equitable public procurement practices.

Type
Commentary
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2025. Published by Cambridge University Press

Policy Significance Statement

The increasing mistrust in technology platforms and emerging technologies, such artificial intelligence, highlights a critical challenge for policymakers. With technology increasingly used in electoral processes, there is a need for policies that ensure transparency, accountability, and trustworthiness of digital platforms to prevent misinformation, manipulation, and mistrust. Recommendations are proposed on how policymakers should work to establish guidelines or frameworks to regulate technology in elections to protect democratic integrity.

1. Introduction

As artificial intelligence (AI) and other emerging technologies continue to dominate policy discourse and debate about how these technologies would shape different aspects of society, there are expressions of cautious optimism and concern about the role of these technologies. For the conduct of elections, a particular area of concern is the role of AI in spreading election misinformation and the effect on the integrity of elections. A dominant consequential effect of AI is the ease of generation of deepfakes and fake news to create narratives that can influence the outcome of elections. This has exposed the fragility of democracy across the globe and resulted in policy proposals to tackle these AI-enabled propaganda (Effoduh, Reference Effoduh2024). This development has also created what is described as the “liar’s dividend,” a phenomenon where politicians also falsely claim that unfavourable stories about them are fake news or deepfakes (Schiff et al., Reference Schiff, Schiff and Bueno2023).

These complexities have led to other impacts of AI and other new technologies on elections being ignored. This article looks at the effect of emerging technologies on the conduct of elections, how their application has been contested, and the governance choices that need to be made in Africa to enable a more positive application of AI and other emerging technologies in Africa.

2. Electoral practices in Africa and the consequent inefficiencies

Before the colonialisation of sub-Saharan Africa, African societies had a wide variety of polities with widely different forms of government varying from powerful centralised monarchies or empires to the so-called stateless societies in which some form of public order was maintained without any centralised institution of the type which Western observers were able to classify as constituting a state. The idea of popular sovereignty expressed through elections is in many cases, newer (Ellis et al., Reference Ellis and Abbink2000).

When colonial powers occupied Africa in the 19th century, they imposed local rulers on many societies. For societies that had chiefs, they either displaced the existing rulers and imposed rulers of their own, or forged alliances with the incumbent chiefs (Bates, Reference Bates2010). After World War II, the imperialists were no longer able to dominate the course of events outside their European base and, therefore, sought to change the outcomes they could no longer control. Where in the past, they could pick or impose local political leaders, they now had to be satisfied with merely shaping the way the leaders were chosen. Consequently, out of political necessity, they began to introduce their own version of representative institutions and displaced pre-colonial representative institutions. They permitted prominent locals to take legislative and executive counsel roles and, after some interval, allowed local citizens to choose who among them were to do so. The culminating process was “self-government,” the assumption of full executive and legislative power by local leaders (Bates, Reference Bates2010).

The colonial governments may have been forced to introduce electoral competition and representative institutions, but the device they employed to shape these institutions was not one that pursued democratic rights and freedoms but one that sought to safeguard the institution that had secured foreign domination over the African people (Bates, Reference Bates2010).

Twenty-six sub-Saharan countries in Africa had gained independence by the late 1960s. However, the general optimism that a new world was being created from the best mix of African and European traditions felt at the independence of many African countries was quickly tempered by the realities of political life. By 1960, when most of Africa became independent, nine countries had formed one-party regimes. By the mid-1970s, seven more governments imposed single-party rule. In the initial period just before and after independence, the formation of opposition tended to be voluntary or the product of negotiation. Political consolidation became involuntary and a product of rigged elections, jailing of political opponents, and outlawing of political parties. This was followed by a wave of coups and the formation of military regimes. In some cases, the military chose to hand back power to civilians, but in others, they remained in office. By the mid-1990s, authoritarian regimes had become a dominant feature of African politics (Bates, Reference Bates2010).

The pattern of democratic process differs among African countries as with the nature, context, and outcome of elections. The way political transition was negotiated influenced the electoral processes. Four patterns emerged: first was countries where civil society was clamorous and organised, took the initiative for political reforms, engaged the state fiercely, and stamped its will on the course of the transition. The outcomes of this were that the new political actors ensured that electoral laws were overhauled to allow for free and fair electoral competition and the institutional infrastructure of elections. The second category was countries where civil society took the initiative for political reforms that were scuttled by the ruling regime and, consequently, the existing restrictive electoral laws were either left untouched or were not significantly altered and the authoritarian control of the regime remained intact. Third was where the state took the initiative for political reforms promoting a form of guided democracy in which it consciously managed, regulated, and manipulated the electoral process to impose its interests and little was achieved in terms of meaningful results through elections. The fourth was political transitions that devolved into severe political conflicts and civil wars (Adejumobi, Reference Adejumobi2000). However, in Africa’s recent political history, there has been a new wave of democratisation backed by new constitutions promoting electoral integrity, but as the case studies below will demonstrate, the outcomes of these constitutional developments are being sabotaged by both state and non-state actors.

These trends suggest that elections are only an expedient political exercise for ruling regimes, partly because of their economic implications in terms of external aid flows and economic assistance, and partly because of their public relations advantage in polishing the political profile of the regime. All these trends have resulted in a deep mistrust for elections in Africa by African citizens and in a bid to promote the legitimacy of the outcome of these organised political events called elections, various technologies have been introduced and their application have further eroded trust in Africa’s fragile democracy.

3. Use of technology in Africa’s elections: the intention and outcome

Election processes depend on trust for their legitimacy. The basic tasks of election administration—voter registration, records and content management, chains of custody, and outreach—are all labour-intensive assignments that can be simplified with technology if used effectively. Consequently, various technologies are being deployed to, for example, automatically transfer data from handwritten paper records to build in efficiencies in security, time, and accuracy. Furthermore, programmes are being implemented to capture election results on voting days and prevent the manipulation of election results. In Nigeria’s 2019 elections, to reduce the occurrence of violence and fraud, the Independent National Electoral Commission introduced an automated fingerprint identification system to verify the authenticity of voters, which limited the extent to which electoral fraud could be committed (Chukwuma, Reference Chukwuma2022). Further, in Ghana, facial recognition technology was used to verify voters and prevent impersonation (Ahmed and Maru, Reference Ahmed and Maru2024).

There have been several use cases of adopting technologies for elections in Africa with risks for voter disenfranchisement and highly harmful outcomes. The Independent Electoral Commission (IEC) of South Africa piloted new voter management devices in the 2021 local government elections, calling it the country’s “most technologically advanced election” (Maseko, Reference Maseko2024). These internet-enabled devices were used to electronically capture voter information, ensure voters were in the correct districts, and provide a live tracking tool for voter participation on Election Day. However, the IEC reported that over 100,000 voters’ details were not uploaded into the electoral system, preventing them from casting their votes (Maseko, Reference Maseko2024). While the new technology was largely successful, challenges such as voter exclusion, poor internet connectivity at some polling stations, and the high cost of implementation highlight the ongoing hurdles in adopting election technology across African countries.

During the 2015 Nigerian elections, it was reported that Cambridge Analytica tested its infamous AI-driven data analysis technology as the percussor to its use in the 2016 US elections by hacking the private and personal information of the main opposition candidate (Mavedzenge and Saki 2023). Cambridge Analytica was also reported to rely on AI-embedded social media algorithms to identify and target online supporters of the opposition with disinformation as part of an effort to sway the voting preferences in favour of the re-election of the President (Mavedzenge and Saki 2023).

While the use of technology in Africa’s elections have been intentional as part of political strategies to sway election outcomes with the manipulative use of technologies, there have also been some positive intentionality on the part of some African governments to strengthen electoral processes and improve the integrity of the outcomes. With the results of Africa’s elections often disputed with allegations of vote rigging, a rationale for the adoption of emerging technologies in conducting elections is to build back trust in electoral processes.

In 2022, Kenya held its third general election since the adoption of the 2010 Constitution. This was an important milestone for Kenya, given the tragedy of the 2008 elections, when several lives were lost following the violence disputing the outcome of the 2008 elections. An important outcome of that tragic episode was the adoption of the 2010 Constitution, which provided important electoral reform in Kenya. This was to address the lack of trust in the electoral system that had endured in Kenya for a long time. As part of the reform, Section 44 of the Elections Act of 2011 introduced the use of electoral technology. The Act required the electoral body to adopt technology in the electoral process and the Independent Electoral and Boundaries Commission (IEBC) subsequently developed a technology known as the Kenya Integrated Electoral Management System (KIEMS), which utilised a hybrid approach of both technology and manual processes for elections. The KIEMS was used for biometric voter registration and, on the Election Day, for voter identification, as well as the transmission of election results from polling stations to the National Tallying Centre.

Following the declaration of President Ruto as the winner of the 2022 elections with <51% of the votes, Raila Odinga, the main opposition presidential candidate who lost the election with over 48% of votes, challenged the outcome of the election on a few grounds including the technology used by IEBC during the 2022 election. First, it was submitted to the Supreme Court that the way the technology was deployed and used fell short of prescribed constitutional and statutory standards (Odinga vs. Ruto, 2022). Second, it was submitted that the IEBC violated its constitutional duty by delegating the design, implementation, and conduct of the KIEMS component of the election to a foreign company—Smartmatic International Holding BV (Odinga vs. Ruto, 2022). Third, Odinga argued that the IEBC vigorously fought any attempt to subject Smartmatic’s activities to accountability and transparency, including the safeguards required by the Elections Act and its regulations (Odinga vs. Ruto, 2022).

The arguments presented by Odinga are a broader reflection of the mistrust that the use of technologies in Africa’s elections are fuelling. In an environment where public institutions are deeply perceived to be untrustworthy, this further raises the imperative for the development of technologies that are safe and trustworthy. Such development requires consideration not only for the technology design but also for the relationship between the state and the citizen, as well as between the citizen and industry.

In response to Odinga’s arguments to the Supreme Court, the IEBC submitted that the electoral system met the constitutional threshold; that all necessary information was accessed only by authorised persons; the information was accurate, complete, and protected from malicious modification either by authorised or unauthorised persons; it maintained an audit trail on activities related to information; and the information was available and could be authenticated using various security features (Odinga vs. Ruto, 2022).

The Supreme Court did not engage substantively with the arguments presented by Odinga and simply focussed on the evidential burden on the applicants and rejected the arguments by Odinga on the basis that credible evidence was not presented for the allegations made to support the required standard of proof.

While the Kenyan Supreme Court decision did not find credible evidence of the allegations made, from a public perception perspective, a shadow of doubt was cast on the integrity of the conduct of the elections and the ensuing outcome. Consequently, while the original rationale for the adoption of technologies was to build back trust, often, allegations made about the deployment and use of those very same technologies serve to further erode trust in electoral processes.

The distrust in Africa’s elections is not only a result of the allegations made by opposition parties when they lose elections, but also through the misuse of emerging technologies by other actors during and after elections. While social media platforms such as WhatsApp, Facebook, and X have been a valuable tool for citizen self-reporting about events happening in their voting stations, social media platforms have also been weaponised to spread misinformation to drive specific narratives. The rise of AI has been central to these objectives. For example, the rise of AI has been used to influence public opinion by spreading misinformation or disinformation. In 2023, there was a leaked audio that turned out to be a deepfake widely distributed over WhatsApp, Facebook, and other social media platforms during Nigeria’s 2023 general elections. The fake audio purported to depict a secret conversation between leaders of Nigeria’s main opposition party, the People’s Democratic Party, planning to compromise elections. This rise of sophisticated deepfake technologies and massive language models raises sociotechnical problems about elections.

At a broader level, the deployment of technologies in Africa needs to be considered in relation to their impact on the civil and political rights of citizens in Africa. It is also important to consider infrastructural inequalities impact at both the individual and state levels (Breckenridge, Reference Breckenridge2021). In Kenya and other developing countries where electricity supply may be unstable, this means essential systems may go offline: the Kenya elections were in fact threatened by the introduction of new electoral systems that struggled with both electricity supply and Internet coverage, which would have disenfranchised millions of voters without the ability to rely on a manual backup for voting (Privacy International, 2013). The deployment of technologies in running elections are thus both developmental and implicate rights concerns that need to be addressed.

4. What the future holds for technology in Africa’s elections

There are three important implications that should be considered for the adoption of various technologies in running elections in Africa. The first is the adoption of a human rights-based approach, and the second is the role of private actors in conducting elections. Third, it is necessary to ensure the adoption of technologies does not heighten digital inequalities.

4.1. Human rights implications

A human rights-based approach to the application and governance of technologies is necessary because new technologies such as AI have become an intractable challenge facing many states given its complexity in terms of governance and its application to society. Juxtaposing the potential of AI to advance innovation with its potential for harm raises questions on accountability on the design, deployment, and use of AI and the fundamental protection of human rights.

The application of human rights as a framework to assess the impact of AI is profound because it prevents an assessment of AI technology in the abstract, and the evaluation is organised around a nearly universally accepted human rights standard recognised by governments and the international community.

Notwithstanding, human rights frameworks are only as good as their ratification by countries, which makes enforcement difficult. With AI governance still based on the implementation of voluntary principles and self-regulation, an important consideration becomes what suitable oversight mechanisms should be deployed over the use of AI to ensure accountability. While governments hold the primary responsibility to uphold human rights and the rule of law, private sector obligations in the protection of human rights in AI-based systems is paramount and a state’s own duty in this area can be difficult to perform, such as where the nature and impact of AI-based systems are not well understood. For example, in the Kenya case discussed earlier, the opposition challenged the use of a foreign company to conduct the elections and accused the electoral commission of outsourcing its constitutional responsibilities to the private company. The transnational profile of technology developers and the cross-border application of the systems highlight the difficulty in regulation where international law and extraterritorial application of national laws may be needed to address impacts felt in different countries.

Technological developments are fast outpacing legal developments and the ability of lawmakers and regulators to keep up shows the need for states to think of more agile, flexible, and networked regulation for a human rights-based approach to technology governance. The primary outcome that countries are seeking in a human rights-based approach to technology governance is to centre accountability in the system’s lifecycle, and that starts from how these technologies are developed or procured in the first place.

According to the Global Index on Responsible AI (GIRAI), an index that measures the extent to which state and non-state actors are establishing and implementing frameworks and protecting and promoting human rights in the context of AI across 138 countries, AI governance remains an idea rather than a concrete approach (GIRAI Report, Reference Adams, Adeleke, Florido, de Magalhães Santos, Grossman, Junck and Stone2024). However, few countries such as Brazil in its AI strategy focus on the importance of AI impact assessments to measure the real and potential harm of AI systems, “access to redress and remedy where harm occurs, and public procurement guidelines that address the adoption of AI by the public sector which oftentimes includes the use of AI in the delivery of socio-economic rights and services to citizens” (GIRAI Report, 2024). This human rights impact assessment approach is needed for state actors to conduct participatory, transparent, and meaningful reviews for election technologies that are adopted for electoral cycles.

4.2. Public–private partnerships

The political intersection of public sector interests with private sector interests should be the underscoring political economy used to consider incentives and risks within emerging technologies, given the dominance of public–private partnerships (albeit of different types) that dominate the African landscape. The use of technologies developed by the private sector for a public service, such as in conducting elections, has implications for public procurement. Public procurement plays a significant role in driving the adoption of new technologies, particularly in the public sector, and improving public procurement laws and policies will have a significant impact on the development and deployment of new and emerging technologies.

Throughout the evolution of technologies, governments have played a central role as regulators and facilitators mandated to create enabling environments for industries to thrive (Kuziemski and Misuraca, Reference Kuziemski and Misuraca2020). Moreover, governments have formed partnerships but also procured technologies in the quest to deliver for their people. It is not a one-size-fits-all approach, but governments have adopted different approaches to partnership building and procurement agreements. For instance, through multi-stakeholder partnerships bringing together non-profit organisations, the private sector, researchers, and innovators, governments can move away from trying to reinvent the wheel in creating new technologies or starting from scratch but rather work with private actors.

Globally, governments are the largest purchaser of technology, which happens via public sector-led procurement processes. For new domains like AI, the public sector needs to adapt existing procurement practices to responsibly acquire and deploy technological tools that improve citizens’ lives while minimising risks. It is projected that governments and public administration will continue to invest in procuring AI to improve public sector services (Rial, Reference Rial, Dimand, Dignum and Dotan2023). Although investments of this size are not evenly observed around the world, evidence suggests governments are increasingly purchasing AI systems, even where there are gaps in procurement practices. These gaps may be caused by the lack of understanding or awareness of a fast-evolving AI landscape, the lack of appropriate tools (i.e., contract templates and decision frameworks), ineffective regulation and guidance required to advance specific AI use cases, and non-collaboration across the public sector, but particularly between procurement officials and policymakers (McGilpin, Reference McGilpin2023).

One of the key objectives of public procurement policies is to promote the responsible and ethical use of technologies within the public sector. This can be done by putting in place strong, transparent, and responsible procurement processes to acquire technology systems. In addition, public procurement policies can be used to promote diversity and competition. This may involve setting aside contracts for small- and medium-sized enterprises and start-ups, promoting the use of open standards and interoperable technologies, and encouraging collaboration between government agencies and the private sector for the design of safe and trustworthy technologies.

Part of the challenges that arise by state dependencies on private sector technologies—which come with risks like vendor lock-in or deferred accountability—may arise because of insufficient capacities in the public sector to design and build these technologies (Breckenridge, Reference Breckenridge2018). Public sector innovation can be a powerful tool for transformative technologies, yet technology in Africa is largely understood as the domain of the private sector (Rodrik, Reference Rodrik2019).

In addition, basic capacities are required not just for independent development, but for more effective public–private collaborations. Open-source software, code, and development can offer the public sector opportunities to avoid vendor lock-in and improve sustainability, but this is only possible in contexts where the in-house capacity exists to maintain (and further develop as needs change) the form of technologies that may be elected. A lack of capacity also has an often underappreciated impact—public procurement processes in most countries have rigid tender specification processes and so forth—but without sufficient in-house technological capacity, the “imaginary” of the public sector to construct solutions through innovation are themselves constrained. This creates a heavy burden on private sector or development partners to try and imbue those innovation processes within partnerships. Consequently, a central strategy in the technology policy of states is capacity strengthening of its employees for the effective assessment and, in some cases, design and deployment of relevant technologies.

An emerging example of this is in Chile and Rwanda, where the national AI policies aim to modernise public procurement processes to ensure effective acquisition and implementation of AI systems in the public sector, including training public officials to improve the efficiency and effectiveness of AI procurement. Consequently, the Chilean government has issued “standard formats for bidding on algorithms and AI projects, which request that suppliers use models with statistical equity metrics, propose additional data protection measures, and conduct bias analyses, among other ethical requirements” (GIRAI Report, 2024). In Rwanda, in the government’s attempt to prevent vendor lock-in, it aims to build capacity and policy tools to engage local AI solution providers through innovation-friendly procurement processes, organise training sessions, invest in hackathons, prizes and challenges to open opportunities for responsible AI applications in the public sector, and establish a risk-sharing fund to support research and development in the public sector (Rwanda National AI Policy, 2022).

4.3. Tackling technological inequalities

Other academic literature are increasingly centralising inequalities not just as an understanding in income disparity, but also across social, political, and technological spheres as a challenge to traditional narratives. In the digital space, these inequalities can be used to frame unequal access to infrastructure and the Internet. Access to these infrastructures is central to contextualising how the impacts of technologies, both beneficial and harmful, may have differential outcomes. Government’s technology policy needs to ensure that the deployment of technologies for elections does not disenfranchise voters and a hybrid approach of adopting new technologies with manual processes is necessary as governments build out the availability of digital infrastructure.

Further, in tackling technological inequalities and engendering trust in the use of technologies in elections, addressing digital skills and literacy among the public is crucial. This is important in tackling misinformation about the impact and role of technologies adopted in elections and the implications of disinformation in disenfranchising voters. A dedicated information campaign and literacy programme during election cycles can address this. Second, investments in digital public infrastructure in elections is crucial and can include safeguarding the integrity of data being used in assessing voter behaviour, safeguarding personal information, and ensuring that third parties only have access to aggregated and open data for research and election monitoring programmes.

5. Conclusion

This article has highlighted how frequently the political economy of technology projects in the public sector is dominated by the development being facilitated by the private sector (sometimes with ties to the development community). The contexts in which these technologies are implemented mean that policy and political dynamics have an exceptional influence on both the priority areas selected for technologies, such as elections, the design of those technologies, and, particularly, their implementation. A human rights-based approach in the governance of technologies is necessary and an agile, networked approach to governance needs to apply when industry partners are transnational. Investments in local innovation can be an important tool in designing contextually adaptable technologies where digital infrastructures are still lacking and technology policies can facilitate an enabling environment for local hubs of innovation to grow. The gaps in public procurement policies further need to be addressed and the capacity of government officials in the assessment of these technologies need to be strengthened.

Ultimately, the use of technologies in Africa’s elections can only be successful in building trust when governments are intentional about building public knowledge in the function and use of these technologies. This requires transparency in government approaches towards the rationale, adoption, and goals of adopting these technologies. Public education cannot be relegated as a mandate of a single institution in government and a “whole-of-government” approach is needed if the adoption of new technologies will serve as a tool for building back trust in Africa’s elections.

Data availability statement

The data that support the findings of this article are from other academic sources that have been appropriately cited and are available in the references section of this article.

Author contribution

Conceptualization: F.A.; Methodology: F.A.; Writing – original draft: F.A.; Writing – review & editing: F.A.

Funding statement

This work received no specific grant from any funding agency, commercial, or not-for-profit sectors.

Competing interests

The author declares none.

References

Adams, R, Adeleke, F, Florido, A, de Magalhães Santos, L G, Grossman, N, Junck, L and Stone, K (2024) Global Index on Responsible AI Report, 1st Edn. South Africa: Global Center on AI Governance.Google Scholar
Adejumobi, S (2000) Elections in Africa: A fading shadow of democracy?. International Political Science Review 21(1), 5973CrossRefGoogle Scholar
Ahmed, S and Maru, M (2024) AI can Make African Elections More Efficient – But Trust Must be Built and Proper Rules Put in Place. Available at https://theconversation.com/ai-can-make-african-elections-more-efficient-but-trust-must-be-built-and-proper-rules-put-in-place-229841 (Accessed April 10, 2025)Google Scholar
Bates, RH (2010) Democracy in Africa: A very short history. From Impunity to Accountability: Africa’s Development in the 21st Century 77(4), 11331148.Google Scholar
Breckenridge, K (2018) The Global Ambitions of the Biometric Anti-Bank: Net1, Lockin and the Technologies of African Financialisation International Review of Applied Economics 33(1), 93118.CrossRefGoogle Scholar
Breckenridge, K (2021) What happened to the theory of African Capitalism? Economy and Society 50(1), 935.CrossRefGoogle Scholar
Chukwuma, I (2022) The role of ICT in curtailing electoral fraud and violence in Nigeria Africa Development. Africa Development 47(2), 161178.Google Scholar
Effoduh, J (2024) The Role and Potential of Artificial Intelligence in Extremist Fuelled Election Misinformation in Africa. Available at https://gnet-research.org/2024/03/08/the-role-and-potential-of-artificial-intelligence-in-extremist-fuelled-election-misinformation-in-africa/.CrossRefGoogle Scholar
Ellis, S (2000) Elections in Africa in historical context. In Abbink, GJ (ed.), Election Observation and Democratization in Africa. Basingstoke: Macmillan, pp. 3749.CrossRefGoogle Scholar
Kuziemski, M and Misuraca, G (2020) ‘AI governance in the public sector: Three tales from the frontiers of automated decision-making in democratic settings, Telecommunications Policy 44(6), 113.CrossRefGoogle ScholarPubMed
Maseko, M (2024) Effects of Digital Technologies on Africa’s Electoral Democracy: Perspectives from South Africa. Digital Policy Studies 3(1), 4457.CrossRefGoogle Scholar
Mavedzenge & Saki (2023) The possible future of African democracy in the age of artificial intelligence In Ncube, et al, Artificial Intelligence and the Law in Africa LexisNexis. Lexis Nexis: South AfricaGoogle Scholar
McGilpin, B (2023) ‘New Recommendations Released for Procurement of AI Tools in Government’. Available at https://blog.thegovlab.org/new-recommendations-released-for-procurement-of-ai-tools-in-government.Google Scholar
Ministry of ICT and Innovation, Rwanda (2022) The National AI Policy. Available at https://rura.rw/fileadmin/Documents/ICT/Laws/Rwanda_national_Artificial_intelligence_Policy.pdfGoogle Scholar
Odinga & 16 others v Ruto & 10 others (2022) Law Society of Kenya & 4 others (Amicus Curiae) (Presidential Election Petition E005, E001, E002, E003, E004, E007 & E008 of 2022 (Consolidated) KESC 54 (KLR) (Election Petitions).Google Scholar
Privacy International (2013) ‘Biometrics: Friend or Foe of Privacy?’ Briefing. Available at https://privacyinternational.org/news analysis/1409/biometrics-friend-or-foe-privacy.Google Scholar
Rial, E, Dimand, AM, Dignum, V and Dotan, R (2023) How to manage AI procurement in public administration. Available at https://www.weforum.org/stories/2023/07/how-to-manage-ai-procurement-in-public-administration/.Google Scholar
Rodrik, D (2019) ‘Public or Private: What’s Really Driving Technological Innovation?’ World Economic Forum. Available at https://www.weforum.org/agenda/2020/08/democratizing-innovation/.Google Scholar
Schiff, KJ, Schiff, DS and Bueno, N (2023) “The Liar’s Dividend: The Impact of Deepfakes and Fake News on Trust in Political Discourse,” SocArXiv x43ph, Center for Open Science, USA.Google Scholar
Submit a response

Comments

No Comments have been published for this article.