How Generative AI From IBM Watsonx Boosts Productivity at the US Open

By Noah Syken, VP of sports and entertainment partnerships at IBMI'm old enough to remember when mullets were cool — not vintage or ironic, but legitimately cool. I also remember the guy who made them cool: a flashy, talented tennis star from Las Vegas, Nevada, named Andre Agassi.This year marks the 30th anniversary of Agassi's first US Open Championship, which is around the time that IBM first partnered with the US Tennis Association to build and maintain its digital operations. So, we asked Andre to help us articulate just how much has changed in tennis and technology since then.

IBM is working with the USTA to refine and improve the already world-class digital experience of the 2024 US Open app and website — a task we've taken seriously for the last 30 years. This year, we're introducing a new feature called Match Reports, which uses generative AI models built with IBM watsonx to help produce long-form, narrative summaries of every singles match during the two weeks of the tournament.

The technology behind this content engine is impressive, especially the use of large language models, including IBM Granite . But this is really a story about people — specifically, the USTA editorial team that develops the articles found on the US Open app and website. I've gotten to know many of the talented and hardworking folks on this team over the years.

But there are limits to the amount of coverage this team of writers and editors can produce. The US Open is a sprawling tournament with 22 courts, hundreds of matches, and countless storylines. It would take an army of reporters working around the clock to cover it all. But now, generative AI takes the first step, producing a cogent, comprehensive summary of a match – complete with scores, key statistics, and momentum shifts – within minutes of completion. The USTA editors then review and edit the content, adding color and context, before approving and publishing.

"In the past, we would have to pick and choose which matches were going to have the best stories, and that's a tough thing to predict from an editorial perspective," said Brian Ryerson, senior director of digital strategy at the USTA. "With our Match Reports, we'll be able to have full coverage of every singles match during the main draw."

To me, this is the productivity promise of generative AI. The USTA use case demonstrates an AI-powered workflow that scales capability. And while AI gets a lot of credit in the headlines these days, it's still the hard work of people like Brian Ryerson and his team that makes these systems work.

This brings me back to Andre Agassi. Early in his career, Agassi's athletic ability, outrageous hair, and flashy clothes garnered headlines and made him an instant style icon. But behind the flair was a man who worked his tail off from a young age, practicing and grinding his way to eight major championships – including two US Opens and an Olympic gold medal. It is a good reminder that even as fashion changes and technology advances, success is ultimately built from the hard work of human beings.

This post was created by IBM with Insider Studios .

How generative AI from IBM watsonx is supercharging productivity at the US Open