Fans at the Olympics are trying out a new AI-powered talent identification system designed to identify future gold medallists. The developers aim to make this technology portable to bring advanced sports science to remote regions globally.
As the alarm sounds, Tacto frantically swipes at infrared sensors that flash blue. Nearby, his younger brother Tomo races down a short track, with cameras tracking his movements.
The seven- and four-year-old siblings from Yokohama, Japan, are participating in AI-driven tests set up near the Olympic Stadium in Paris to discover potential future champions. The system evaluates data from five tests, including running, jumping, and grip strength, to assess power, explosiveness, endurance, reaction time, strength, and agility. This data is then compared to that of professional and Olympic athletes.
Sarah Vickers, head of Intel’s Olympic and Paralympic Program, explains that the system uses computer vision and historical data to allow individuals to compare themselves with elite athletes and determine which sport they are most suited for. After completing the tests, participants receive recommendations for sports they might excel in from a list of ten.
Intel assures that all collected data is deleted after the process. Despite the technology, the young brothers are having a great time. “I enjoyed it,” says Tacto. “I liked the sprinting part the most.”
The AI system available to fans at Paris 2024 has a more portable version that can run on basic devices like mobile phones, tablets, or PCs. “With just a mobile phone or tablet, you can reach places previously inaccessible,” says Sarah. This portable AI technology assesses performance by analyzing video from a camera, eliminating the need for physical sensors.
The International Olympic Committee recently used this system in Senegal, assessing over 1,000 children in five villages. Partnering with Senegal’s National Olympic Committee, the system identified 48 children with significant potential and one with exceptional potential. These children have been invited to join sports programs to further develop their abilities.
The goal is to extend the system’s reach, providing opportunities in locations difficult to access with traditional, bulkier assessment tools. Prof. John Brewer from the University of Suffolk notes that identifying young talent is crucial but warns that a basic system measuring a few attributes may not capture the full potential for technical sports or endurance events. “You need to assess aerobic capacity, which filming alone cannot show,” he says.
Nevertheless, Prof. Brewer acknowledges the system’s value for initial assessments. “If it reveals skill and agility suggesting potential in a sport, it should be encouraged,” he says. “And if it’s portable, bringing it to areas lacking high-tech assessments is a positive step.”
At the Olympic Stadium, young Tacto learns he has potential as a sprinter, though he currently prefers football and tennis. Meanwhile, experienced athletes Hank and Brock, who competed in college-level sports in the US, find the technology intriguing. “This kind of technology wasn’t around when we were competing,” Hank says.
Their results? “Rugby for me,” says Hank. “Basketball for me, and I’ve never played it,” replies Brock. “Well, he played once with me and we never let him back,” Hank laughs.
It seems even with AI, technology can sometimes miss the mark.
