SYRACUSE, N.Y. (NCC News) — The Winter Olympics in Milan concluded on Feb. 22, with the U.S. bringing home 12 gold medals and 33 in total, the largest medal haul by Team USA since the 2010 Winter Olympics in Vancouver. However, as many of Team USA’s coaches would say, the medals were not won in a single day. They were the result of countless hours spent conditioning, practicing, analyzing, recovering and studying.
Nothing matches blood, sweat and tears when it comes to preparing for the Olympics, but the 2026 games were a stage where AI showed huge potential in changing how athletes and coaches practice and perform.
In figure skating, AI helped create 3D models of jumps that give coaches the ability to analyze technique and form in real time. In alpine skiing, drones and specialized sensors analyzed skiers’ poses through turns, giving coaches enough time to notify the racer in the starting gate of any condition that could influence the next run.
Olympic sports require the observation tools and methods to be dynamic, since they include time-based, judgement-based and competition sports. The games proved these tools can be adapted to virtually any sport through the use of cameras and drones.
Hassan Rafique, assistant professor of sport analytics at the Falk College of Sport and Human Dynamics at Syracuse University, studies how machine learning can advance sports analysis in several ways.
“In sport-specific contexts, analysts use specialized third-party platforms that provide domain expertise, such as wind and strategy analytics in sailing, possession and expected-goals models in ice hockey, or spatial-tactical analysis tools,” said Rafique.
Cameras are the most important tool for the kind of data collection that happens during sport events. AI vision models paired with drones that fly over athletes in competition analyze the athlete’s pose, and analysts are then able to extract insights from the data.
So, the same analysis that was available at the Olympics with the use of cameras can be used in the most mainstream sports in America. The process relies on maneuverable cameras located above the playing surface, close enough to have a view of whatever player is being analyzed.
So, could we end up seeing AI-data models working during professional and collegiate sports? Rafique says there is a real possibility.
“I could see some of the technology like the AI vision models being used,” said Rafique, “Those allow us to estimate the pose of athletes, so for example if you’re a basketball player, we can analyze shooting form and estimate the trajectory of the ball.”
The data that is collected can be significant to fans, but could also end up being very valuable to coaches.
Syracuse University club lacrosse coach Jon Bang thinks that the data could help his players better understand the specific mechanics of good form.
“I think sometimes, the players need to see it executed perfectly by someone,” said Bang. “A lot of times that can be better than just saying ‘hey this is what we should be doing.'”
The possibility of having live AI analysis of player data somewhere like Syracuse is closer than most might think, and could be at most major sporting events soon. To athletes, fans, and coaches it could be the next step in understanding preparation and performance.