3 Questions: Using AI to help Olympic skaters land a quint

2026-02-10 · Source: MIT News - Artificial intelligence · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics · Depth: Intermediate, medium

Summary

MIT Sports Lab researchers Jerry Lu and Anette "Peko" Hosoi are applying AI to enhance figure skating performance and analyze aesthetic evaluation. Lu developed OOFSkate, an optical tracking system that uses AI to analyze video of skaters' jumps, providing physical metrics and comparisons to elite athletes to improve technical elements like quadruple axels and Salchows. This system also classifies trick execution to approximate international judging scores. Hosoi is researching how AI systems evaluate aesthetic performance, collaborating with Professor Arthur Bahr and Eric Liu to understand if AI reasoning pathways for aesthetic judgment align with human experts or novices. Lu will also work with NBC Sports for the 2026 Winter Olympics, using AI to explain complex scoring in figure skating, snowboarding, and skiing.

Key takeaway

For AI Scientists developing models for complex human activities, this research highlights how sports like figure skating offer rich, quantifiable data for both technical and aesthetic evaluation. You should consider using such domains to fine-tune AI models, as understanding human judgment in these contexts can significantly advance artificial general intelligence research and model development.

Key insights

AI can objectively analyze figure skating technical elements and explore subjective aesthetic judgment.

Principles

• Pose estimators excel where depth data is less critical.
• Aesthetic evaluation data exists in judged sports.
• AI research benefits from understanding human judgment.

Method

OOFSkate uses video analysis and AI to track physical jump metrics, compare them to elite athletes, and provide an automated grade of execution score, aiding technical improvement.

In practice

• Use OOFSkate for technical jump analysis.
• Compare your jump data to Olympic champions.
• Apply AI pose estimators in 2D-dominant sports.

Topics

• AI in Sports
• Figure Skating Analytics
• Optical Tracking Systems
• Pose Estimation
• Aesthetic AI Evaluation

Best for: Computer Vision Engineer, AI Scientist, AI Researcher, AI Engineer, Research Scientist

Related on AIssential

• 3 Questions: Using AI to help Olympic skaters land a quint — AI can objectively analyze technical sports performance and explore subjective aesthetic evaluation.
• Efficiently Linking Real Scenes with Synthetic Data Generation for AI-based Cognitive Robotics and Computer Vision Applications — Bridging the sim-to-real domain gap via linked synthetic and real data is crucial for scalable AI robotics.
• How Google Cloud is helping Team USA elevate their tricks with AI — AI-powered video analysis transforms smartphones into biomechanics labs for real-time athletic feedback.
• Runway started by helping filmmakers — now it wants to beat Google at AI — Training AI on observational video data and world models, not just language, is the next frontier for advanced intelligence.
• AI is coming to Olympic judging: what makes it a game changer? — AI in Olympic judging promises consistency and transparency but faces challenges with trust, cultural values, and subjective elements.
• Can you *really* train AI to "get" videos just by showing it a million of them? — Current video models excel at generation but lack systematic evaluation for genuine spatiotemporal reasoning.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by MIT News - Artificial intelligence.