Wimbledon features hundreds of matches played across 18 courts. Keeping up with all the action can be a challenge, even for the most avid fans. So Wimbledon and IBM collaborated to develop a new feature called Catch Me Up, that serves up succinct, personalised summaries designed to get fans quickly up to speed on their favourite players and the major storylines from around the tournament, including recent performances and upcoming matches.

The Catch Me Up feature is produced by a generative AI model built with IBM watsonx, that is fueled by Wimbledon’s trusted data sources, and has been trained to describe the game of tennis and Wimbledon in particular. The summaries are personalised to a fan’s favourite players and appear on the home page of the Wimbledon app and website.

To achieve this, the IBM Consulting^® team built a generative AI solution based on IBM Granite, a family of powerful large language models (LLM) that are purpose-built for business and engineered from scratch for scalability to build trust in AI-driven applications. The team chose a model called “Granite 13b chat,” which boasts 13 billion parameters and is optimised for dialogue use cases such as virtual agents or chatbots.

The teams uses the IBM watsonx AI and data platform to manage the lifecycle of the AI model, from curating trusted data sources to governing responsible, trusted AI. Work began with IBM^® watsonx.data^™, a data store that connects disparate sources and allows developers to filter the data for things like profanity, hate speech or personally identifiable information. For Catch Me Up, the team processed nearly 3 terabytes of data.

This data was used to train the Granite LLM within IBM watsonx.ai^™, a next-generation studio for building and training generative AI models for business use cases. The IBM team then added specific domain expertise of Wimbledon, including the use of unique Wimbledon nomenclature such as “gentlemen’s draw” rather than “men’s draw.” The team monitors the ongoing performance of the AI using governance tools from IBM watsonx.governance^™.

As a result, the Wimbledon editorial team are able to scale content production to reach both new and existing tennis fans globally, as well as provide fans access to timelier, curated coverage across all ladies’ and gentlemen’s singles matches, personalised to their preferences.

To develop new capabilities like Catch Me Up every year, the AELTC needs to move with speed and purpose. That requires combining the right people, processes and technology to facilitate rapid innovation.

It starts with a long-standing collaboration between the Wimbledon team and the experts at IBM Consulting. “We work together to find alignment on areas of interest, generating a combined vision for how fans will interact with our products in the future,” says Clements. “The combined passion in delivering experiences of the highest quality supercharges our partnership and ensures we can make the most of the opportunities in front of us.”

The teams use the IBM Garage^™ Method to foster creativity and gain a deep understanding of tennis fans. As part of their biannual innovation workshops, the Wimbledon and IBM teams seek to address two key challenges: How to engage fans more deeply with Wimbledon and tennis on an ongoing basis—for many, The Championships is the only tennis event they watch all year—and how to encourage fans to follow up-and-coming players who aren’t famous yet.

Through these workshops, the team develops concepts and features for the app and website. For example, SlamTracker has been the premier scoring application for Wimbledon for many years, providing real-time scores, statistics, and in-depth, point-by-point analyses of every match. This year, SlamTracker has been completely redesigned. It now features generative AI-produced Match Previews, including “Likelihood to Win,” a head-to-head match prediction generated from a comprehensive analysis including player statistics, media sentiment, and player momentum.

In order to transform these new ideas into digital reality, IBM Consulting built a platform of innovation for Wimbledon, capable of processing structured and unstructured data, and integrating technology from a variety of sources. Automated workflows integrate and orchestrate the flow of data through the various applications and AI models needed to produce the digital experience.

These workflows are made possible by a hybrid cloud architecture and containerised apps running on Red Hat^® OpenShift^®. To keep the entire operation running smoothly, the team uses IBM Instana^®Observability technology, which monitors application performance and surfaces issues requiring resolution, so the team can take swift action and avoid any downtime.

And this year, the team took advantage of a powerful generative AI assistant to increase the efficiency and accuracy of its code. IBM watsonx Code Assistant uses generative AI to accelerate software development, helping developers generate code based on natural language prompts. The team used this tool to analyse and explain snippets of code, annotate code to facilitate better collaboration between developers, and auto-complete snippets of analysed code.

Finally, IBM is responsible for the security of the entire Wimbledon digital platform, charged with safeguarding it from cyber attacks, protecting sensitive data, and managing access to the system. Wimbledon is a high profile event and a tempting target for hackers and cyberattacks. The experts from IBM Security handle threat management, including detection and response, using a variety of tools available through IBM’s Managed Security Services offering.

Known simply as “Wimbledon,” The Championships is the oldest of tennis’s four Grand Slams and one of the world’s highest-profile sporting events. Organised by the AELTC (link resides outside of IBM) and based in London, Wimbledon has been a global sporting and cultural institution since 1877.