3 Questions: Using AI to help Olympic skaters land a quint | MIT News
3 Questions: Using AI to help Olympic skaters land a quint | MIT News
http://newsoffice.mit.edu/2026/3-questions-using-ai-help-olympic-skaters-land-quint-0210
Publish Date: 2026-02-10 00:00:00
Source Domain: newsoffice.mit.edu
- Jerry Lu MFin ’24 developed OOFSkate, an optical tracking system using AI to analyze figure skaters’ jumps and provide performance improvement recommendations.
- Lu worked with Team USA’s elite skaters to enhance their technical performance and will collaborate with NBC Sports during the 2026 Winter Olympics to explain complex scoring systems for figure skating, snowboarding, and skiing.
- Professor Anette Hosoi and Lu discussed the potential of AI in assessing aesthetic performance in figure skating, which is inherently subjective and different from the technical aspects that are easier to measure.
- The MIT Sports Lab is researching whether AI systems in figure skating could evaluate the aesthetic side of the sport, specifically looking into how these systems arrive at their judgments and if they mimic human reasoning.
- The article explores the potential for AI to impact the understanding of how humans, experts, and novices judge aesthetically, with skating serving as a useful model due to the availability of data.
- The researchers are optimistic that a skater will land a quint (five revolutions in an Axel jump) eventually soon, but the human physical limits likely prevent higher jumps like a six-rotation move.
- Olympic figure skating, with its difficulty disguised by the elegance of skating, offers a unique opportunity for AI to showcase the sport’s underlying complexity and to make it more relatable to viewers.
