This year’s Wimbledon comes with an extra dollop of AI spin: generative AI commentary on matches and a powerful, live draw analysis tool.
For now, the AI commentary will be packaged with match highlights. Deciding the key moments to include was already guided by AI, based on factors such as crowd reactions, but in 2023 it will also feature AI-generated audio commentary.
This, however, is just the start. In the future, Wimbledon viewers may be able to hear live, AI-generated commentaries on all matches. Not just highlights, not just matches that feature big names.
Once this year’s draw has been made, Wimbledon app users will also see a draw predictor based upon years’ worth of data. Think head-to-head matchups, past performance on grass, recent form and how strong the competition is in each players’ part of the draw.
How the Wimbledon AI commentary works
Kevin Farrar, IBM UK & Ireland Sports Partnerships Leader, explained that the AI commentary will be based on IBM watsonx, the company’s AI and data platform for enterprise.
“So we have taken one of the IBM foundation models and we have trained it in the language of tennis and the language of Wimbledon,” said Farrar at a briefing to announce the new tools, just a stone’s throw from Centre Court.
“And then we are using generative AI to create a narration, a commentary on the AI highlights reels that we produce for the men’s and ladies’ singles matches,” he added. “This is a stepping stone on a journey where ultimately we would like to be able to generate a commentary on full matches.”
The commentary is not based on a specific commentator but an amalgam of hundreds. “We have tennis specialists on the team, so we’ve drawn on them in terms of the language that is used,” said Farrar. “It’s not based on an individual and their style.”
But that isn’t to say such a thing couldn’t happen. “You could see in the future, you could train it in different styles… and it opens up all the possibilities around different languages, different voices.”
So, an AI version of John McEnroe AI on call for any match? We cannot believe it!
Introducing the Wimbledon predictor tool
IBM and Wimbledon’s second big announcement was the IBM AI Draw Analysis tool.
“[This] is the first statistics of its kind in tennis,” said Chris Clements, Digital Products Lead at The All England Club. “A 128-player draw has many potential matchups, and it’s hard to see how the draw is shaping up in the early stages. This statistic uses AI to define how favourable the path to the final might be for each player in the singles draw.”
Clements drew a comparison with the current seeding system, which is simply based on how players have performed in the past year. “A player’s favourability rating on the IBM AI Draw Analysis, however, is based on a player’s potential future matchups against potential future opponents.
“It also factors in the overall quadrant difficulty or where the player is placed in the draw compared to other players. For example, a highly seeded player could receive a bad draw if they go up against a surprisingly difficult lower ranked opponent.”
The idea behind the tool is to educate and engage fans, many of whom will only watch tennis once a year: Wimbledon.
“This new insight will help tennis fans uncover anomalies and potential surprises in the draw, which will not be apparent by looking only at players’ seeds and tour ranking,” said Clements.
“In turn, we expect that to inspire more debate and engagement in the fan community.”
The Wimbledon Championships run from Monday 3 July to Sunday 16 July, with the draw taking place shortly beforehand.
To see the AI technology in action, visit wimbledon.com or download the Wimbledon App (available from the App Store and Google Play Store).
Nathalie Parent, Chief People Officer at Shift Technology: “HR is the conscience of an organisation”
For more than 30 years, Nathalie Parent has led global HR teams, working primarily with software companies. Today she’s Chief People Officer at Shift Technology
Amazon introduces new storage class that makes it cheaper to store rarely used files
Robot carers are real, but caregiving has bigger problems, writes Richard Trenholm in this FlashForward edition