City of Cape Town moves to counter AI-generated fake news posts

The City of Cape Town is stepping up efforts to deal with fake news and manipulated content circulating online.

Advances in artificial intelligence have made it easier to produce convincing false posts and images.

There's a lot of it around, says CapeTalk host John Maytham, noting that some is easier to spot than others.

City officials have recently begun watermarking official videos and photographs, a move aimed at helping residents quickly identify authentic City communications.

Alderman JP Smith, City of Cape Town Mayoral Committee Member for Safety and Security, says the watermarks are being added to his own social media posts.

"Because what happens very quickly is that your content is repurposed as scams. There are conmen who will try to repurpose your account."

Catching AI-generated fake news is particularly important as local election campaigning approaches, he says.

"[It] serves a political or social or ideological narrative of whoever sends it, which is not your exact content but adjusted in such a way that it speaks to the message somebody is trying to communicate."

Smith says the spread of AI-generated misinformation is not unique to Cape Town. The issue was widely discussed at a recent international municipal conference in Canada, where representatives from 270 cities shared similar concerns.

"One of the first examples that was discussed at the conference is the phenomenon of incels and young men who feel alienated and then find spaces on the internet where that alienation is exploited by others and radicalised."

According to Smith, fake posts often originate from anonymous social media accounts or politically motivated actors.

"It's interesting to see how other cities around the world adapt to this, develop tools and set up structures to combat this within their municipalities. Some of that for me was eye-opening."

To listen to Smith in conversation with Cape Talk's John Maytham, use the audio player below: