Chukchansi Park image via Ben Hensley.

ChatGPT did not write this story.

But artificial intelligence (AI) writing stories is not far off, and in the sports world, it may have seen its beginnings right here in Fresno.

The Fresno Grizzlies produce a postgame write up after each of their games. Writeups typically include final game scores, attendance and track important historical statistics, both team and individual.

Writing a postgame story, however, also takes time; and time is something that can be lacking with a team of three to five game-day staff members, each given a plethora of tasks each and every game, including operating the newly-implemented pitch clock, scorekeeping, datacasting (creating the play-by-play gamecast data through MiLB.com), and running social media accounts during the game.

Enter ChatGPT.

First experimented with by the Grizzlies earlier this year on Twitter, ChatGPT looks to step in and take a huge load off the shoulders of the game-day staff at Chukchansi Park while still maintaining the quality of the related content.

By utilizing AI learning technology, ChatGPT is not only able to process information and produce a postgame writeup, but even uses terminology and game-specific verbiage to create what used to take front office staff hours to create themselves.

Stephen Rice, Grizzlies Media Relations Coordinator and Play-by-Play Broadcaster who has written postgame reports for the Grizzlies since 2016, was initially skeptical about the system, but since testing, praises it for its efficiency and potential to save time for front office staff.

“After looking at how quickly it will save time and it still doesn’t change my tone of voice – it still allows for a mix of my voice and also what we’re looking for – I said ‘why not,’” said Rice. “I know that there’s enough portfolio to not only my stuff but Jonathan [Bravo] and the Grizzlies stuff that we can allow it to do this and it won’t take away from the total product.”

Despite initially not knowing specific teams or players it has not been introduced to, ChatGPT is able to learn from everything inputted into its interface, allowing Rice or other Grizzlies media members to copy/paste the postgame play-by-play into the system.

That allows the program to learn and remember team statistics throughout the game and eventually, throughout the season.

After ChatGPT receives the information it takes no longer than 30 seconds to a minute for the program to fully construct a 400 to 800-word story. After that, the staff will spend roughly 15 minutes to fact check, edit and publish the postgame recap, dropping what used to take hours to a matter of, in most cases, no longer than 20 minutes.

“It’s harder to describe than it is to show,” Rice said. “But knowing that we have the system and the stuff in place to implement it correctly makes us feel so much better about the whole process.”

As of now, both Rice and Grizzlies Director of Marketing and Communications Jonathan Bravo believe the Grizzlies to be the only team in professional sports that plan to utilize ChatGPT in their upcoming season’s reporting.

Rice said that during a typical Tuesday through Sunday six-game homestand, he arrives at the ballpark between 7 a.m. and 8 a.m., often not leaving until 1 a.m.

It is his hope that ChatGPT will not only allow himself and others to be more productive at the ballpark, but will also provide the team with more recovery time during a season that sees 132 games played in just 155 days.

“Even though it’s something that we still need in this industry, if it takes less time for it, it’s going to be a big factor,” Rice said. “If it means that I get home at 11 p.m. instead of 1 a.m., I’m not going to be too upset about that.”