The Looming Threat of AI-Powered Disinformation: A Global Challenge with Acute African Implications
The rapid advancement of artificial intelligence (AI) has ushered in an era of unprecedented technological possibilities, but it has also opened the door to a new wave of disinformation, easily and cheaply produced, with potentially devastating consequences. Experts warn that the ability to create realistic deepfakes – videos and audios manipulated to depict events that never happened – poses a significant threat to democratic processes, social cohesion, and international stability. Hendrik Sittig, director of the Konrad Adenauer Foundation’s media program in sub-Saharan Africa, encapsulates the pervasive anxiety, stating that the potential ramifications of AI-generated disinformation in the coming years are "really scary." This concern is echoed by the World Economic Forum’s 2024 Global Risk Report, which identifies AI-supported disinformation as a top global threat, primarily aimed at undermining democratic principles and fracturing societies.
A recent study conducted by the Konrad Adenauer Foundation, the Institute for Security Studies in South Africa, and the cyberintelligence institute in Germany examined the proliferation of AI-driven disinformation in Africa and Europe, focusing on its deployment around national elections. The research reveals that these campaigns are frequently designed to erode public trust in electoral authorities and processes, sowing doubt about the legitimacy of elections. Notably, the study highlights the under-researched nature of AI disinformation beyond the electoral context in Africa, underscoring the need for further investigation into its broader societal impact. The findings reveal striking similarities in the challenges faced by both continents, with similar AI tools being employed by various actors, particularly far-right political parties, to manipulate public opinion and spread propaganda.
The Global Actors Behind AI Disinformation and Their African Targets
While Russia has emerged as a leading proponent of AI-generated disinformation as a foreign policy tool, other actors, including those from China and the Gulf states, are also actively targeting Africa to disseminate their narratives. These actors leverage AI to generate compelling, yet fabricated, content designed to sway public opinion and advance their interests. Christopher Nehring, from the cyberintelligence institute in Germany, points to a range of actors, including groups linked to foreign states, cybercriminals, terrorist organizations, and Islamist groups, who utilize AI to create and spread disinformation online. This finding aligns with previous research on the topic.
Interestingly, the study reveals that despite the potent threat of deepfakes, their dissemination in Africa remains relatively limited compared to Europe. This is attributed to the high cost of internet and social media access in many African nations, making these platforms inaccessible to a large segment of the population. Consequently, less sophisticated "cheap fakes," which are easier to detect, are more prevalent. Nevertheless, the potential for deepfakes to cause significant disruption remains a serious concern, particularly in politically charged contexts like elections and conflict zones.
Case Studies: Disinformation and Conflict in the DRC, Elections in South Africa and Burkina Faso
The ongoing conflict in the eastern Democratic Republic of Congo serves as a stark example of how disinformation and hate speech, often amplified by AI, can exacerbate tensions and fuel conflict. Karen Allen, from the Institute for Security Studies, notes that images and text content linked to Rwandan accounts have significantly shaped public opinion, reinforcing suspicions and aggravating the conflict. She highlights the tactic of flooding social media platforms with AI-generated content to drown out dissenting voices, a strategy frequently employed in conflict zones and during elections.
South Africa, with its large social media user base, experienced a surge in AI-supported disinformation during the 2024 parliamentary elections. A deepfake video of former US President Donald Trump purportedly endorsing the newly formed Umkhonto we Sizwe party, led by ex-President Jacob Zuma, was widely circulated, becoming the most shared AI-generated content during the election period. This underscores the vulnerability of even relatively technologically advanced African nations to AI-manipulated information. Similarly, following the 2022 coup in Burkina Faso, AI-generated videos featuring avatars posing as Pan-Africanists were disseminated to garner support for the military junta. These videos, suspected to be linked to the Wagner Group, exemplify the sophisticated use of AI tools like Synthesia to create deceptive narratives and manipulate public sentiment.
Combating the Disinformation Deluge: Challenges and Strategies
The proliferation of AI-generated disinformation presents a complex challenge, exacerbated by the decision of major social media platforms like X and Facebook to discontinue their fact-checking programs, shifting the burden of content verification onto individual users. This poses a significant challenge, particularly in regions like Africa where access to diverse and reliable information sources can be limited. Karen Allen emphasizes the importance of accessing information from a "variety of sources" to mitigate the risk of manipulation. While Africa lags behind Europe in terms of robust data protection regulations, increased collaboration on fact-checking practices and platforms for reporting suspected disinformation are crucial first steps. The study also suggests that Africa’s developing regulatory landscape presents an opportunity to learn from the experiences of other regions and implement effective strategies for combating AI-driven disinformation. The fight against AI-powered disinformation demands a multi-pronged approach involving robust regulations, media literacy initiatives, and enhanced international cooperation to counter this growing threat to democracy and societal harmony.