The subject is taken very seriously. Three months before the Olympic and Paralympic Games in Paris (from July 26 to September 8), the French authorities and the International Olympic Committee (IOC) want to ensure “mental safety” athletes in “protecting them from cyberbullying”. Athletes are increasingly the targets of mockery, insults and even death threats on the Internet; these are found both in comments on online press articles or YouTube posts as well as on the social networks X, Instagram, Facebook and TikTok.
“It’s so pervasive and unfortunately normalized that it’s hard to get a sense of the scale of the problemassures World Kirsty Burrows, Head of the Safe Sport Unit at the IOC. If someone attacks you online right before your competition, you won’t feel safe, even if your physical environment is safe. So we must try to support athletes by removing these threats. »
Thus, the institution will use artificial intelligence (AI) – managed by a service provider linked to the platforms – capable of withdrawing “in real time in more than thirty-five languages and dialects” hostile messages from the social accounts of some 15,000 Olympic and Paralympic athletes, as well as those of coaches and officials (more than 2,000 people). “They will all be able to access this system if they wish, we will offer it de facto”specifies Mme Burrows.
This is the first time that such a device will be deployed throughout the Games. The goal is to remove abusive content before it is even read. “We expect almost half a billion messages on social mediaunderlines Kirsty Burrows. If you spend a second on each [dans une configuration de modération 100 % humaine], it would take you sixteen years to process them all. If we neglect digital security, we will fail in our duty. » In the event that an athlete manages to see the comment before it is deleted, there will also be three psychologists present at the athletes’ village.
“Moderation is a game of cat and mouse”
At the same time, the Ministry of Sports indicates that French and international athletes who wish can turn to Bodyguard, a French company specializing in the protection of social network accounts. Already used by the French Tennis Federation and the Professional Football League, Bodyguard mask thanks to an AI, messages deemed toxic – these are then only visible to their author – but also emojis or other pictograms (in the shape of a banana to insult a black athlete, for example). Indeed, to circumvent the algorithms which identify hateful messages, Internet users “reinvent language”explains Charles Cohen, founder and president of Bodyguard. “They understand how the filters work and seek to circumvent them”he adds.
You have 39.51% of this article left to read. The rest is reserved for subscribers.