The US Open is turning more of its live tennis data into an interactive AI experience. IBM and the United States Tennis Association are introducing new and upgraded features across USOpen.org and the US Open app, including AI-generated Serve Quality scores, Key Moments analysis, a personalized Live Updates homepage and an expanded conversational Match Chat.
IBM Brings AI Match Insights and Serve Analytics to US Open
Watching a tennis match increasingly means watching two games at once: the competition on court and the data stream explaining what is happening.
IBM and the United States Tennis Association (USTA) are pushing that second layer further at the 2026 US Open, adding AI-powered tools designed to help fans understand matches in real time rather than simply follow scores.
The partnership, which dates back to 1992, will introduce three major AI enhancements across USOpen.org and the US Open app: a personalized Live Updates experience, an AI-generated Serve Quality metric and expanded Key Moments and Match Chat capabilities.
The changes arrive as sports organizations increasingly use artificial intelligence to turn enormous volumes of live data into information that fans can understand.
An IBM-commissioned global survey conducted by Morning Consult found that 91% of surveyed tennis fans use sports apps during events, while 64% reported high trust in AI-powered sports content.
Those numbers point to an important shift in sports technology. Fans are not simply looking for faster information. They increasingly expect digital platforms to interpret the information for them.
Turning a tennis serve into data
The most technically ambitious new feature is Serve Quality.
The AI-powered metric will be available across all 254 singles matches and uses limb-tracking technology developed with IBM Bob to analyze the mechanics of players’ serves.
The system tracks 21 data points across the player’s body and racquet at a rate of 50 times per second. That includes movements such as wrist position and the transfer of energy from the legs through the torso.
IBM estimates that the system will generate approximately 1.2 billion data points during the tournament.
That raw information is only useful if it can be processed quickly enough to become part of the fan experience. IBM says the live data stream, including measurements related to efficiency, accuracy, consistency and ball toss, is managed through IBM Confluent.
The result is a Serve Quality score generated in near real time.
For casual viewers, the appeal is relatively simple: instead of knowing only that a player hit a fast serve, fans can get additional context about why the serve was effective.
For IBM and the USTA, the bigger technology demonstration is the pipeline connecting computer vision, streaming data, AI analysis and consumer-facing content.
AI explains why a match changed
Another enhancement targets a different problem.
The existing Likelihood to Win feature estimates each player’s probability of winning using current and historical statistics, expert analysis and match momentum.
Key Moments is designed to add the missing explanation.
Rather than simply indicating that a player’s winning probability has changed, the feature summarizes the turning points and momentum shifts that contributed to that change.
That is an increasingly important role for AI in sports media.
Traditional statistics are good at recording events. Generative and analytical AI can potentially connect those events into a narrative: a sequence of break points, a drop in first-serve effectiveness or a change in rally performance that altered the direction of a match.
The distinction is particularly useful for fans who join a match after it has started. Instead of scrolling through dozens of points, they can quickly understand what changed.
Match Chat becomes more conversational
The US Open’s Match Chat is also getting an upgrade.
The interactive feature allows fans to ask questions in natural language and receive answers based on live match information, historical statistics and analysis. Some answers can now incorporate photos and video.
IBM says Match Chat uses watsonx Orchestrate together with a collection of AI agents and specialized models trained around the USTA’s editorial style and tennis terminology.
That architecture reflects a broader enterprise AI trend: rather than relying on one general-purpose model for every task, organizations are increasingly combining specialized models, data sources and AI agents within controlled workflows.
For a sports organization, this can be particularly useful because the AI needs more than factual knowledge. It has to understand tennis terminology, interpret live data and produce answers in a consistent editorial voice.
That combination is becoming a new challenge for AI-powered media platforms.
The rise of the AI sports interface
The US Open is not alone in experimenting with AI-driven fan experiences. Sports leagues, broadcasters and technology companies are increasingly turning match data into personalized feeds, automated narratives, predictive statistics and conversational interfaces.
The competitive question is moving beyond who can collect the most data.
The differentiator is becoming who can turn that data into useful context without overwhelming the fan.
IBM’s approach at the US Open illustrates that shift. Serve Quality converts biomechanical data into an accessible score. Key Moments converts match statistics into an explanation of momentum. Match Chat converts a complex database into a conversational interface.
The common denominator is abstraction.
Fans do not necessarily want to interpret billions of data points. They want the relevant few.
That puts AI in an interesting position within sports media. The technology is not replacing the underlying statistics or editorial storytelling. Instead, it acts as a translation layer between the complexity of live sports data and the way people consume information.
Enterprise implications beyond tennis
The technology stack behind the US Open also offers lessons for enterprises outside sports.
Real-time AI applications require several components to work together: high-volume data ingestion, low-latency processing, specialized models, workflow orchestration and a user interface that can turn outputs into useful decisions.
That architecture has applications in sectors such as financial services, healthcare, manufacturing and retail.
The challenge is particularly relevant to companies adopting AI agents. An AI system is only as useful as the data and workflows surrounding it. Reliable real-time information must reach the right model, while outputs need appropriate controls and context before reaching customers.
The US Open provides a relatively visible demonstration of this architecture because the data is inherently live and the audience is large.
The tournament runs from August 23 through September 13, giving IBM and the USTA several weeks to demonstrate whether these AI experiences can improve how fans understand the competition.
The more significant development, however, is not any individual feature.
It is the emergence of the sports website and mobile app as an AI-native interface—one where fans no longer have to search through statistics to find meaning. They can ask questions, receive explanations and see complex physical performance translated into understandable insights.
As AI becomes embedded across digital products, that model could become increasingly common far beyond the tennis court.
Market Landscape
The sports technology market is moving toward AI-powered personalization, computer vision, real-time analytics and conversational interfaces.
IBM’s US Open deployment combines several of those technologies into one consumer experience. The architecture spans computer-vision-based body tracking, high-volume data streaming, AI analytics, specialized models and agentic workflows.
The broader enterprise AI market is moving in a similar direction. Microsoft, Google, Amazon and Salesforce are all building AI agents and orchestration capabilities designed to connect models with real-time business data and workflows.
The US Open example demonstrates why that integration matters. A model alone does not create a useful sports experience. The system needs access to live match data, historical context, editorial rules and specialized analytics.
For sports organizations, the commercial opportunity extends beyond engagement. AI can create new premium statistics, personalized content and sponsorship opportunities while potentially reducing the manual effort required to produce match narratives.
The long-term competitive advantage may therefore belong to organizations that have both high-quality proprietary data and the infrastructure to turn it into personalized experiences.
Top Insights
- IBM and USTA are adding AI-powered Serve Quality, Key Moments and Match Chat features to make US Open data more personalized and understandable for fans.
- Serve Quality tracks 21 biomechanical data points 50 times per second, potentially generating 1.2 billion data points throughout the tournament.
- Key Moments builds on Likelihood to Win by explaining the match events and momentum changes behind shifts in winning probabilities.
- Match Chat combines live data, historical information, AI agents and specialized models to provide conversational answers in the USTA’s editorial style.
- The US Open demonstrates how enterprises can combine real-time data, specialized AI models and orchestration workflows to build AI-native customer experiences.
Power Tomorrow’s Intelligence — Build It with TechEdgeAI
