IBM watsonx: Agentic AI Reinvents The Masters Tournament
IBM has unveiled advanced AI-powered digital features for the 90th Masters Tournament, showcasing how enterprise technology platforms could revolutionise fan engagement in live sporting events. The deployment, built on the IBM watsonx platform, introduces agentic AI capabilities that allow users to interact with decades of tournament data and real-time course information through natural language interfaces.
The technology rollout marks 30 years of partnership between IBM and the Masters Tournament, with this year’s innovations centred on making vast datasets immediately accessible and actionable.
Jonathan Adashek, Senior Vice President of Marketing and Communications at IBM, says: “The Masters Tournament and IBM have continually raised the bar on unique digital experiences that blend cutting-edge technology with the timelessness of Augusta National Golf Club.”
The implementation demonstrates how IBM’s enterprise AI solutions, typically deployed across financial services, healthcare and manufacturing sectors, could be adapted for consumer-facing applications requiring real-time data processing and historical archive management.
Natural language search capabilities
The centrepiece of IBM’s 2026 deployment is the Masters Vault Search, a system that transforms more than 50 years of final-round broadcast archives into a searchable digital library. The tool utilises IBM’s Granite small language models and the watsonx Orchestrate platform, enabling users to locate specific moments through conversational prompts rather than manual browsing.
According to IBM, the system employs optical character recognition, speech-to-text transcription and scene detection to index historical metadata. The searchable database includes stroke data extending back to 2015 and tournament results from 1968 onwards.
By deploying multiple AI agents working in concert, the platform can process natural language queries and return relevant video segments from decades of archived content.
The technology addresses a common challenge in media asset management: making historical content discoverable without requiring extensive manual tagging or metadata creation. The approach could have applications across broadcasting, corporate training archives and digital media libraries where legacy content remains largely inaccessible due to indexing limitations.
Real-time probability calculations
IBM has expanded its Hole Insights feature, now entering its third year of operation. The system uses generative AI to calculate real-time scoring probabilities for every shot during tournament play.
Development of the feature involved collaboration with Jim “Bones” Mackay, the former caddie and current commentator, whose course knowledge was integrated with historical performance data.
When a ball comes to rest on the course, its precise coordinates are captured and compared against historical data for that specific hole location. The watsonx platform then calculates the probability of achieving an eagle, birdie, par, bogey or worse, generating contextual insights based on Augusta National's unique terrain and conditions.
The real-time analysis capability demonstrates how AI systems can process live data streams, compare them against historical patterns and generate probabilistic forecasts within seconds.
Jonathan says: “The introduction of Masters Vault Search and updates to Hole Insights show how generative and agentic AI can transform vast amounts of data into meaningful insights – whether you're a golf fan who wants to understand the implications of a single shot in real time or a financial institution using AI to analyse millions of transactions to identify patterns and inform decisions.”
Enterprise applications beyond sport
IBM’s sports and entertainment partnerships, including agreements with Scuderia Ferrari HP, Wimbledon, the US Open and the Grammys, function as public demonstrations of the company’s hybrid cloud and AI solutions.
The same watsonx platform and Granite language models deployed at Augusta National are available to enterprise clients across industries for applications including customer service automation, document analysis and predictive analytics.
The Masters deployment could provide insights into how small language models perform in production environments with millions of concurrent users, real-time data requirements and integration with legacy broadcast systems. These technical challenges mirror those faced by organisations implementing AI across existing infrastructure.
The 90th Masters Tournament started yesterday and will finish on Sunday, with AI-powered features accessible via the official Masters website, mobile app and Apple Vision Pro application.
