South Africa’s New Disinformation Code: A Critical Defence as Public Confidence in Spotting Fake News Collapses
South Africa is heading into what is likely to be its most fiercely contested local government election on 4 November, and the Electoral Commission of South Africa (IEC) has moved to protect the integrity of the vote with a new legally binding code against disinformation. The Electoral Code of Conduct Countering Disinformation in the Local Government Elections, gazetted on 4 September, arrives at a moment of profound vulnerability. According to the IEC’s latest Voter Participation Survey, public confidence in the ability to identify fake news has fallen sharply, from 62% in 2023 to just 52% in 2025/26. That means almost half of South Africans are unsure whether they can tell truth from falsehood in the political information they encounter. Even more troubling, only 18% of respondents say they regularly verify election-related content before liking, sharing, or reposting it. The other 82% admit they rarely, if ever, take that basic precaution. With only six weeks left before voting begins, these figures expose a dangerous gap between the scale of the disinformation threat and the public’s ability to counteract it. The new code, which applies to political parties, candidates, and other electoral stakeholders, carries the full legal weight of the core Electoral Code of Conduct, meaning violations can lead to sanctions. This is not a symbolic gesture; it is an acknowledgement that electoral manipulation has evolved beyond simple rumour-mongering and now demands a structured legal response.
The IEC’s survey reveals a complex and shifting information landscape. Traditional media remain the most trusted and heavily used sources for electoral information, with 71% of respondents relying on television and 38% on radio. Yet social media is surging, with 23% now citing it as a source of election-related information, a dramatic rise from just 8% in 2021. While it has not yet overtaken traditional outlets, South Africans are among the heaviest internet users in the world, spending an average of nine hours and 37 minutes online each day in 2025, the highest recorded globally. Roughly three and a half of those hours are dedicated to social media platforms such as Facebook, X, and WhatsApp. This extensive online presence should be an opportunity for democratic engagement, but it has not translated into stronger verification practices. The disconnect is especially visible in community-based platforms, where WhatsApp groups and local Facebook pages have become the primary arenas for ward-level political discussion. In these spaces, information spreads rapidly through closed networks, often outside the reach of journalists, fact-checkers, or platform moderators. A believable but entirely fabricated message about a candidate’s criminal record, or a doctored audio clip portraying a municipal official accepting a bribe, can circulate among hundreds of voters within minutes. By the time it is debunked, the damage may already be done. The combination of high usage, low verification, and fast, unmoderated sharing creates a fertile environment for disinformation to flourish, particularly when messages are tailored to local issues such as housing, water supply, and service delivery.
The threat is becoming increasingly hyper-local and technologically sophisticated. At the Disinformation Dialogue 2026, IEC Chairperson Mosotho Moepya warned that the upcoming elections would likely see a shift from national misinformation campaigns to “ward-specific deceptions” driven by generative artificial intelligence and highly targeted local campaigning. This warning aligns with research from the Institute for Security Studies (ISS), which shows that disinformation and political messaging are now being localised to an unprecedented degree, aimed at specific communities, language groups, and even individual voting districts. Community WhatsApp groups and local Facebook pages have become the central axes of this strategy, serving as spaces where political agents can insert fabricated content that resonates with local grievances. Generative AI has made the problem far worse: tools that create realistic audio, video, and written text are now cheap, accessible, and difficult to detect. Although South Africa has not yet experienced a major election crisis caused by deepfakes, the potential is real and growing. The World Economic Forum’s Global Risks Report 2026 ranks misinformation and disinformation among the top risks facing societies worldwide, not only because of their immediate impact but because they erode trust in institutions, facts, and democratic processes themselves. In the South African context, where municipal elections are already strained by service delivery protests and political factionalism, the arrival of hyper-local, AI-generated falsehoods could be especially destabilising, turning already tense communities into battlegrounds of confusion and mistrust.
Political parties have not been slow to exploit these new tools. ISS research shows that online political advertising and influencer-driven campaigning are becoming central features of election strategies, with parties investing heavily in digital teams and paid content. These tactics can enhance political engagement by enabling direct dialogue between candidates and voters, but they also blur the boundaries between authentic political conversation, paid propaganda, and coordinated influence operations. A voter scrolling through their feed may struggle to tell whether a post supporting a candidate is a genuine neighbourly opinion, a sponsored advertisement, or part of an orchestrated campaign using fake accounts. Influencers complicate the picture further, as they may be paid to endorse parties without revealing the arrangement. This opacity is deeply corrosive. At the same time, AI-generated content is advancing faster than the public’s ability to scrutinise it, making it increasingly difficult to distinguish between a real video and a synthetic one. While South Africa has not yet experienced a major deepfake election crisis, the cost and technical barriers for creating convincing fake content have collapsed, meaning any party or individual with a smartphone and a few rand can potentially produce false evidence aimed at damaging opponents or deceiving voters. As a result, information integrity is no longer a matter of media ethics alone; it is now an integral part of citizenship, democracy, and governance. Without a shared baseline of reliable, verifiable information, the very concept of voter choice begins to unravel, as decisions are made on the basis of lies and manipulated realities.
Addressing this challenge requires a broad, multi-stakeholder response, and the IEC’s new code is only one component. While much public discussion focuses on social media users, verification cannot be left solely to citizens. Professional and legacy media journalists remain a crucial line of defence. They investigate claims, provide context, and test competing narratives before they reach the public, acting as guardians against falsehood. The Disinformation Code explicitly recognizes this, but media outlets themselves require support and resources to sustain such coverage. Social media platforms also have a significant responsibility. As political parties and candidates are subjected to greater accountability under the code, platforms should strengthen their own responses to election-related disinformation, including fact-checking partnerships, clearer labelling of sponsored political content, and rapid removal of harmful deepfakes. IEC Chairperson Mosotho Moepya has stressed that breaches of the Disinformation Code carry the same legal consequences as violations of the core Electoral Code of Conduct, a strong deterrent if effectively enforced. Yet enforcement is the key test. The IEC must investigate complaints promptly and impose penalties where violations are found, and parties must implement internal controls to restrain their officials, members, and influencers from spreading disinformation. Without concrete examples of parties being held accountable, the code risks being seen as toothless. Furthermore, the code’s temporal limit – it remains in force only until election results are declared – is a weakness, as disinformation can persist long after the polls close. The allegations of vote rigging following the 2024 elections demonstrated that the damage can continue for months, undermining public confidence in the outcome and fuelling divisions.
Therefore, South Africa needs long-term information integrity solutions that extend far beyond election cycles. This includes sustained investments in media literacy programmes in schools and communities, continued support for independent fact-checking organisations, and ongoing research into emerging disinformation tactics. It also requires a cultural shift toward collective vigilance. Safeguarding local democracy is a shared responsibility, but citizens are an important first line of defence. Small, deliberate actions – such as challenging a suspicious post in a community WhatsApp group or helping a family member verify a candidate’s claim – can strengthen democratic resilience far more than any single regulation. The IEC survey’s finding that only 18% of people regularly verify information is both a warning and a call to action. If that figure can be improved, the impact of disinformation will diminish accordingly. With six weeks left before the 4 November vote, South Africans need access to information they can trust. That requires a sustained commitment from political parties, institutions, media organisations, platforms, and citizens themselves. The introduction of the Disinformation Code is a crucial and timely step, but it is only the beginning of a longer struggle to protect the integrity of information upon which democracy depends. The question is not whether South Africa has the tools to fight disinformation, but whether it has the will to use them consistently and effectively once the cameras fade and the election is over. [Written by Adivhaho Ramaite, Research Officer, Executive Director’s office, ISS Pretoria]
