Think for a moment about what it feels like to be lied to before an election. You are scrolling through social media and a short video appears of a local candidate apparently caught saying something outrageous. Your heart pounds. You forward it to your family group. Later you learn that the video was doctored and the quote was fabricated. That moment of confusion and anger is exactly what disinformation wants to create. It does not simply change one person’s mind; it eats away at trust in the whole process, making people wonder whether their vote matters at all. With South Africa’s local government elections scheduled for 4 November, the Electoral Commission (IEC) has taken an important step to fight this threat. It has gazetted a code of conduct designed specifically to counter disinformation, misinformation and fake news. The timing is crucial, because election periods are when false information explodes. Rival campaigns want to gain an advantage, and the internet makes it cheap and easy to spread lies. The IEC itself has identified disinformation as among the biggest risks to the integrity of elections and to the legitimacy of elected governments. That is not an exaggeration. When citizens cannot agree on basic facts, democracy itself becomes fragile. The code is meant to provide guardrails for political parties, candidates and everyone who acts on their behalf. It does not ban criticism or negative campaigning. It does not tell people what to believe. What it does is demand honesty in how election communication is carried out, so that voters are not treated as targets in a manipulation game. It is a reminder that the right to vote is meaningless if the information we use to decide is built on lies. Fake stories can cause real harm: people may be tricked into going to the wrong polling station, persuaded not to vote, or convinced that their ballot will not count. The new code is an effort to stop those harms before they become a permanent part of our political culture.
To understand why a code of this kind is necessary, it helps to sort through the vocabulary. Disinformation is deliberately false information, created and shared with the intention to deceive. Misinformation is false information that may be shared without malicious intent, by someone who genuinely believes it is true. Fake news is a looser term, often used to describe fabricated stories dressed up as journalism. All three are dangerous, especially in the online spaces where many South Africans now get their news. Social media platforms are not neutral word-of-mouth; their algorithms reward engagement, and outrage is the most engaging emotion. A wild accusation can travel around a community in minutes, while a careful correction is ignored. The rise of artificial intelligence makes the problem even heavier. AI-generated images, videos and voice clones can put words into a politician’s mouth or show events that never happened. This is no longer science fiction. Free apps allow an ordinary person to create a convincing fake in less than an hour. For voters, the effect is corrosive. People begin to feel that there is no truth, only spin. Some disengage and stay away from the polls. Others are manipulated into voting against their own interests. The IEC’s approach is not to police the internet generally; it is to hold those who have chosen to run for office to a higher standard. The code requires them to be careful, transparent and accountable in everything they say and share, online or offline. That is a significant commitment, and it is a welcome one. It recognises that in a digital society, the health of democracy depends not only on what happens inside voting stations but on the quality of the public conversation that leads up to election day. If that conversation is poisoned with manufactured lies, then no matter how well the voting machines work, the will of the people is distorted.
The scope of the code is deliberately broad and practical. It applies not only to political parties and candidates themselves, but also to anyone acting on their instruction, authority or for reward. That includes campaign managers, media strategists, social media influencers who are paid to post, volunteers who run fan pages, and even people who share content for a party because they expect to benefit in some way. The code reaches all election-related communication, whether it happens online or offline, paid or unpaid. A speech at a rally, a leaflet dropped in a postbox, a tweet, a WhatsApp message, a Facebook post, or a paid advertisement on Instagram all fall under the same rules. This is important because modern campaigns have many hands. A candidate cannot simply say “I did not post that” if the post was created by an agent to support the campaign. The party and the candidate are responsible for their campaign ecosystem. The phrase used by the commission is “guardrails”, and that is a helpful way to think about it. The code does not tell parties what to say and does not prevent them from criticising their opponents. It tells them how to speak: with evidence, by testing their claims, and without resorting to fabrication. It also places an obligation on them to take control of content published for them or on their behalf. If a supporter creates a false meme, the candidate cannot simply shrug. They must be willing to look at what is circulating in their name and disavow it if it is untrue. This is a significant shift in responsibility for campaign communication. It recognises that politics in the digital age is no longer a game of single speeches; it is an ongoing flow of content, and every piece of that flow carries the candidate’s fingerprint. The public deserves to know who is really behind the messages they receive, and the code helps make that possible.
The code lays out specific obligations that political parties and candidates must meet. First, they must check information for accuracy before sharing it. This sounds like common sense, but in the heat of a campaign it is not always done. The point is that something cannot be shared simply because it feels true or because it came from a friend. Second, they must take responsibility for content published by them, for them, or on their behalf. If they have retweeted a false story, they own it. Third, if they discover that they have distributed false or misleading information, they must publicly retract and correct it within 36 hours of becoming aware. The time limit is important. A lie that survives for two days in the final stretch of an election can change enough minds to change the outcome. The window is tight enough to force prompt accountability but realistic enough for careful handling. Along with the retraction, they must censure those responsible where appropriate. That means they cannot quietly remove a post and hope nobody notices; they have to address the person who created or amplified the lie. Fourth, they must clearly label AI-generated content before distributing it. If a campaign releases a synthetic image or a deepfake audio clip, even for satirical or artistic purposes, the audience must know that it is not real. This requirement is a powerful safeguard against the newest form of manipulation. Fifth, they must clearly label political advertising. When a voter sees a sponsored post, they should be able to see who paid for it and why. Finally, the code requires them to report instances of online disinformation to Real411 within 36 hours. Real411 is the platform designated by the IEC for reporting and tracking suspicious content. This creates an early warning system, so that false stories can be logged, reviewed and countered before they go viral. The goal is not merely to punish bad actors; it is to stop the infection of the public conversation before it does permanent damage. Taken together, these obligations form a kind of digital hygiene plan for political communication, one that treats a lie as a disease rather than a strategy.
Beyond the positive duties, the code prohibits a range of deceptive practices outright. Fake accounts are banned, as are bot networks. A bot network is a cluster of accounts controlled by software that can make a fake trend look like a mass movement or an unpopular candidate look overwhelmingly popular. Coordinated inauthentic behaviour is also banned. This is a more subtle menace. It happens when groups of people work together to hide their real identities and artificially inflate support, attack opponents or engineer fake controversies. The code also targets doxing, which is publishing private information about someone in order to intimidate them, and cyber bullying, which is using online harassment to silence a political opponent or their supporters. These practices are toxic because they do not merely mislead voters; they drive people out of public life. Some candidates have been forced to withdraw, and some ordinary citizens have become afraid to express their views because of online mobs. The code also specifically prohibits disinformation targeting the IEC or its staff. This might include lies about election officials being corrupt, fake announcements about polling station changes, or claims that ballots have been destroyed. If such disinformation occurs, parties and candidates must correct it publicly. This provision is essential because IEC staff are human beings doing a difficult job under enormous pressure. When they are demonised, they become targets of threats and violence, and the whole election is endangered. By requiring the political class to defend the credibility of the electoral process, the code helps protect the neutral space in which democracy operates. It sends a clear message: attacking the fairness of the machinery is not an acceptable campaign tactic. Online disinformation is not a victimless crime. It can literally determine who is elected and how people treat the results, which is why these rules matter so deeply.
The Disinformation Code is not merely a set of good intentions. The IEC has made it clear that a contravention is subject to the same sanctions as a breach of the Electoral Code of Conduct under the Municipal Electoral Act 27 of 2000. That means real legal consequences, not just a bad headline. The commission has designated Real411 as the formal channel for reports. Any individual, party, candidate or organisation can use it to flag suspected election-related disinformation. This is a powerful form of empowerment for ordinary voters. You do not need to be a digital detective. If you see something that appears false or manipulative, you can send a report and let the system do its work. The public commitment to the code will be made at a signing ceremony on 23 September. That ceremony is more than a photo opportunity; it is a pledge made to voters. But the most important protection against disinformation remains with citizens. We can pause before we hit share. We can check whether a story has been verified. We can ask questions about the source of an image or the account that posted it. We can refuse to be turned into amplifiers of lies. The local government elections on 4 November will determine who serves in our municipalities, who manages budgets, who responds to problems with water, electricity, housing and public safety. These are not abstract political matters. They affect the roads we drive on, the cleanliness of our streets, the reliability of our lights and the quality of the schools our children attend. If disinformation makes us stay home in confusion, or makes us vote for someone because of a lie, the entire community loses. The IEC’s code is a brave and necessary step, but it cannot succeed in isolation. It needs all of us to choose honesty over anger and truth over convenience. In the end, democracy is not only about ballots; it is about trust. Preserving that trust is everyone’s duty.