IBM and the USTA Introduce New AI-Powered Fan Experiences for 2026 US Open

August 24, 2026 | Source: IBM Newsroom AI

Tags: IBM, watsonx, USTA, sports AI, real-time AI

IBM and the USTA debut three AI features for the 2026 US Open: a personalized Live Updates homepage, a Serve Quality score built from 21 body-tracking points at 50 frames/sec (~1.2B data points per tournament), and Key Moments context layered on Likelihood to Win — all powered by watsonx Orchestrate.

Details

IBM and the USTA have expanded their 34-year partnership with three new AI-powered fan features going live during the 2026 US Open. The headline addition is Serve Quality, a near-real-time metric that uses IBM Bob limb-tracking to capture 21 data points across a player's body and racquet at 50 times per second. Across all 254 singles matches, the system is expected to generate roughly 1.2 billion data points managed via IBM Confluent, producing a Serve Quality score covering efficiency, accuracy, consistency, and ball toss. The second feature, Key Moments, layers on top of the existing Likelihood to Win probability model — which uses historical stats, expert opinion, and match momentum — to narrate the turning points that shifted each match. Match Chat, the third update, now serves photo and video responses alongside text, making it a multimodal companion rather than a text-only chatbot powered by watsonx Orchestrate. According to a Morning Consult survey of 20,000+ fans commissioned by IBM, 91% of global tennis fans use sports apps during events and 64% trust AI-powered content. The deployment is essentially a live stress test of watsonx in a high-traffic consumer sports context, reaching over 14 million fans annually since 1992.