Intel-powered AI technology detects potential Olympians

These limitations have led to a significant gap: many talented individuals never get the chance to pursue professional sports careers, and fans miss out on witnessing some of the world’s best athletes.

To address this issue, Intel, the Worldwide AI Platform Partner of the Olympic and Paralympic Games Paris 2024, has teamed up with the International Olympic Committee (IOC) and the National Olympic Committee (NOC) of Senegal to deliver an innovative AI-driven sports talent identification application. This platform is designed to level the playing field by expanding scouts’ reach, particularly in underserved regions.

The AI platform, developed through this collaboration, utilises Intel’s technology to detect potential athletic talent via any device equipped with a camera. This eliminates the need for costly specialised equipment and broadens the opportunity for more individuals to be recognised as future Olympic hopefuls.

The technology operates by using computer vision to analyse video footage, with the AI delivering detailed stats to talent scouts. This process significantly simplifies the scouting process, making it more accessible and efficient.

Caroline Rhoades, Olympic & Paralympic Games Partner Marketing Manager of Intel’s Sales, Marketing and Communications Group, stated: “This AI platform designed for talent identification not only uncovers hidden talent but also helps to bridge an existing gap.”

In March, Intel and the IOC rolled out the application in Senegal, aiming to identify future Olympians among the nation's youth. This initiative was strategically implemented ahead of the Youth Olympic Games Dakar 2026, set to be hosted in Senegal’s capital. Representatives from Intel and the IOC visited six villages across the country, using the talent identification app, powered by Intel’s AI platform, to assess and analyse the physical abilities of over 1,000 children. These assessments included tests of speed, acceleration, burst power, agility, and directional change ability, measured through more than 1,000 biomechanical data points.

The testing process was remarkably efficient, lasting only minutes per participant, yet delivering comprehensive data to talent scouts. As Rhoades noted: “As time passes, the hope is that this technology can help increase the chances for every aspiring athlete to have the opportunity to shine on the global stage.”

To support the app's deployment, Intel utilised additional technologies to ensure the success of physical testing, even in the most remote areas. These technologies included:

• Edge devices running on Intel Xeon processors for video processing, real-time computer vision, and biomechanical data analysis
• Intel Gaudi AI accelerators to enhance the efficiency of model training
• AI models optimised with OpenVINO
• Intel Core Ultra notebooks for real-time inferencing

Intel’s AI platform technology handles the critical task of analysing, scoring, and ranking each video, providing scouts, coaches, and national governing bodies with a robust dataset of raw physical talent to review.

The results of this programme in Senegal were promising: out of the 1,000 children who participated, 40 were identified as having significant athletic potential. This initiative not only highlights the power of AI in sports talent identification but also represents a significant step towards ensuring that no talent goes unnoticed, regardless of geographical barriers.