As generative artificial intelligence becomes increasingly accessible, African democracies face an unprecedented threat. With at least 17 African countries scheduled to hold elections in 2025, following 24 in 2024, the continent stands at a critical juncture where AI-generated propaganda could fundamentally undermine democratic processes already plagued by traditional challenges.
The 2023 Nigerian presidential election marked a turning point, experiencing what was likely the first deluge of AI-generated electoral content on the continent. Fake endorsements from Hollywood celebrities, former U.S. presidents, and prominent Nigerians flooded social media. Most concerning were AI-generated audio clips shared hours before the election, purportedly revealing candidates’ plans to rig the vote. Similar patterns have emerged across Africa since, for example the 2024 elections in South Africa saw AI-generated videos of fake speeches by U.S. President Joe Biden threatening sanctions.
Perhaps most alarmingly, AI propaganda has influenced military coups before, during, and after they occur. A 2018 video of Gabon’s President Ali Bongo, with unnaturally stiff expressions, raised questions about his fitness to rule and contributed to a failed coup attempt in 2019. Following Burkina Faso’s September 2022 coup, AI-generated videos from American Pan-Africanists expressing support for the new military regime circulated widely on social media. Residents within Burkina Faso and across Africa are also regularly subjected to propaganda regarding the new regime, which often depicts Ibrahim Traore making critical speeches in English (a language he has not been verified to speak fluently), and supported in protests through repurposed video clips.
Vulnerability and Challenges
Africa has become increasingly vulnerable to these challenges due to changes outside its borders. Research shows that 60% of disinformation campaigns targeting Africa are foreign-sponsored, primarily from Russia, China, and Gulf states, with West Africa bearing the brunt of these attacks. Meanwhile social media platforms’ safeguards, designed primarily for dominant languages like English, leave gaping holes in protection for African users and others across Global Majority countries.
Another challenge is that although fact-checking organizations like Africa Check, Dubawa in Nigeria, and PesaCheck in Kenya have made significant strides in recent years, they face an uphill battle. Africa Check screens 100,000 inquiries annually, but fact-checking is a manual, time-intensive process that cannot match the speed at which AI-generated content spreads. Currently the various measures employed by social media companies, such as forwarding limits and collaboratively labeling misleading posts, are easily circumvented and often ineffective against sophisticated audiovisual AI-generated content like TikTok videos.
Managing the threat of AI-generated propaganda
Addressing this crisis of AI-generated propaganda within Africa will require coordinated action across four key areas. First, expanding citizen education. Current media literacy programs primarily target journalists and media professionals. Governments and civil society must develop accessible programs training in identifying AI-generated content for the public, particularly targeting those with limited digital literacy.
Second, strengthen regulatory frameworks. While 38 of 54 African countries have enacted data protection laws, comprehensive frameworks specifically addressing generative AI remain lacking. The African Union Malabo Convention on Cyber Security and Personal Data Protection offers a foundation, but only 15 countries have ratified it. More broadly, African governments must establish robust regulatory mechanisms that can keep pace with rapidly evolving AI technologies.
Third, investing in technical solutions. In light of large tech companies like Meta deprioritizing fact-checking, African governments should support research and development of AI detection tools tailored to local languages and cultural contexts. Current AI-assisted fact-checking systems, trained primarily on Western datasets, often fail to understand African cultural nuances and may misidentify legitimate content as disinformation, along with the inverse.
Fourth, fostering continental cooperation: Given that 60% of disinformation campaigns within Africa are foreign-sponsored, African nations must work together to identify and counter these threats. This includes sharing intelligence, coordinating responses, and developing continental standards for AI governance. Relying on regional economic communities such as ECOWAS and others can also help fill in the regulatory gaps for countries that lack the individual capacity to enact protections.
Democratic stability in the age of AI
The preliminary impact of generative AI on recent African elections and coups provides a sobering preview of what lies ahead. As these technologies become more sophisticated and accessible, the threat will only intensify. The technology that promises to revolutionize agriculture, education, and government services could instead become a weapon that undermines the very foundation of democratic governance. The evidence from Burkina Faso, Gabon, Nigeria, Senegal, South Africa, and other countries across the continent demonstrates that AI-generated disinformation is already here, already influencing public opinion, and being weaponized to destabilize governments.
The battle for African democracy in the age of AI could be won or lost in the next few years. Success against this challenge will necessitate unprecedented cooperation between governments, civil society, academia, and technology companies. The cost of inaction, which could result in democratic backsliding, increased political instability, and the erosion of public trust, far exceeds the investment needed to build robust defenses against AI-generated disinformation and propaganda. African leaders must act now to ensure that the promise of AI technology serves to strengthen rather than undermine the continent’s democratic future. The window for proactive action is closing rapidly, and the stakes could not be higher.
Chinasa T. Okolo is a Fellow at the Center for Technology Innovation at the Brookings Institution and a researcher, strategist, and policy advisor on AI Governance and Safety.
