"One trillion dollar problem" - disinformation as a business risk

Düsseldorf. In mid-August, a conspicuous shitstorm breaks out against one of the world's largest watch manufacturers. Swatch has just launched a new advertising campaign. In it, an Asian-looking model pulls her eyes back with her fingers. And the viral fury begins almost in real time.

Swatch has made racist insults, they say, Swatch has placed a disrespectful advert, Swatch must be boycotted. The pressure is growing and growing: the share price of the Swiss companyfalls by just under three per centSwatch has to withdraw its campaign, there is even an official apology. And many in the company are wondering what has actually just happened.

Steffen Konrath, founder of the data analyser evAI, also became suspicious at the time. Sure, advertising is controversial. But such a huge wave of excitement? He begins his analysis. His results show what is wrong: The number of negative online posts about Swatch rose by 24,000 per cent in two days. But it was mainly fake accounts that sparked this outrage. Konrath found that 30 per cent of the profiles participating in the discourse were not accounts of real people, but fakes.

There is a method behind this: "These accounts have systematically reinforced existing resentment," says Konrath. AI bots falsify content, imitate supposedly real people and post automatically on social media. "This can turn into a disaster for companies," says Konrath. Such deepfakes are now increasingly targeting companies - and causing real economic damage.

The perpetrators behind AI campaigns come from many directions at once, confuse customers and consumers and can cause lasting damage to a brand's reputation. The victims themselves are barely prepared for a world in which excitement is used as an automated weapon.