Can AI Judge a Skating Competition?

Apr 24, 2025 | Sports

Our Take

We may need to wait for the ice to harden under this idea; high-quality footage of synchronized skating is relatively scarce, meaning the CNNs will have a limited sample size to learn from. But the use of AI as a “sky judge,” to put it in NFL terms, seems inevitable. On the surface, it’s a simple question: would athletes and fans want to see more fairly and consistently officiated/judged sports? The answer would seem to be a resounding “yes.”

But to expect a transition overnight is wishful thinking. Instead, a process where Chan’s synchrobot learns through exposure, in conjunction with human judges, would be the next logical step. The synchrobot would not replace the judges, but rather would gradually lessen their burden of finding each and every flaw in a routine.

Source Summary

Whether it be enhancing the fan experience or tracking detailed player analytics, AI has done tremendous things for the sports world—but can it judge sports? In synchronized figure skating, one of the most subjective of athletic endeavors, we may soon find out.

In a recent Ted Talk, Team USA figure skater Amelie Chan analyzed the problems with the current scoring system: during synchronized figure skating competitions, there are up to 16 skaters on the ice at once. Many times during a performance athletes split into different move groups, each performing its own element.

AI has the potential to create a more fair and accurate judging system by removing human biases and catching things human judges might miss. A “Synchrobot,” as Chan dubs it, could consistently analyze skaters by measuring aspects like body line and leg angles. The form of AI best suited to perform the analysis? Convolutional neural networks (CNNs), which can analyze and classify images to compare skaters’ movements to specified judging criteria.

Can AI Judge a Skating Competition?

Our Take

Source Summary

Forces of Change