Your Tennis Swing Needs Work. This AI Startup Wants to Help [CNET]

View Article on CNET

SportAI uses computer vision, machine learning and biometrics for technique analysis.

Lisa joined CNET after more than 20 years as a reporter and editor. Career highlights include a 2020 story about problematic brand mascots, which preceded historic name changes, and going viral in 2021 after daring to ask, “Why are cans of cranberry sauce labeled upside-down?” She has interviewed celebrities like Serena Williams, Brian Cox and Tracee Ellis Ross. Anna Kendrick said her name sounds like a character from Beverly Hills, 90210. Rick Astley asked if she knew what Rickrolling was. She lives outside Atlanta with her son, two golden retrievers and two cats.

Expertise Technology | AI | Advertising | Retail

Norwegian chess player Magnus Carlsen, ranked No. 1 in the world, was the second-youngest chess grandmaster in history in 2004 at the age of 13.

At 33, he’s backing an Oslo-based SportAI, which seeks to provide commentary and analysis for other sports, first and foremost tennis. (Even though we don’t see chess competition at the Olympics, the International Olympic Committee recognizes chess as a sport.)

AI Atlas art badge tag

The AI sports technology company hopes to democratize access to technique analysis in an AI-influenced sports market projected to reach $30 billion by 2032. Add that to the way artificial intelligence — not just chatbots like OpenAI’s ChatGPT but also things like computer vision — is influencing exercise, art, gardening and other hobbies and activities by generating new content for us and suggesting ideas we haven’t tried before. 

AI-powered swing tips

If you were to take a tennis lesson with a human coach and ask for advice about your swing, the help you get would likely vary based on the coach and his or her background and training.

“Today, technique analysis within sports is very subjective, expensive and unscalable,” said CEO Lauren Pedersen.

The founders of SportAI, which include former NCAA tennis player Pedersen, saw an opportunity to analyze and track player technique on a more personal level using computer vision, which is the field of AI that allows computers to analyze visual content the way humans do; machine learning, which uses data and algorithms to enable AI to learn the way we do; and biometric analysis, which is the measurement of our physical and behavioral characteristics.

After analyzing a video from mobile phones, tennis court cameras or broadcast feeds, SportAI can compare your technique to any number of tennis players, including the best in the world, one particular player you admire, another player you want to emulate or the thousands of players the model saw during training. Thereafter, SportAI is able to give personalized feedback about what you need to do to improve. (SportAI also trained its model using input from coaches to understand what distinguishes good technique.)

In a demo video of a tennis player, SportAI gauges the curve of the player’s swing compared to a pro’s curve and highlights areas for improvement. Then, it measures hip and shoulder rotation, as well as swing velocity and suggests the player transfer her weight to the direction she wants to hit. It also rates her angle of attack and curve compared to pros and recommends she opt for a compact backswing with a solid follow-through to improve her swing.

The startup is beginning with racquet sports like tennis and paddleball, but that’s mostly because of the team’s personal expertise. SportAI plans to expand into other sports with swings, like golf, baseball and cricket.

But this isn’t a product you’ll be able to buy as a consumer. Instead, SportAI is positioning itself as a business-to-business platform and is looking to partner with coaching associations, training academies, retailers, brands and broadcast media.

In the latter case, it would supplement live sports commentary and “engage fans and show how [Spanish tennis pro Rafael] Nadal’s forehand or serve compares to [Swiss former pro Roger] Federer, for example,” Pedersen said.

One of its first clients is a training platform that operates tennis clubs. (Pedersen did not disclose the name.)

“These clubs have cameras mounted at the back of the court, so as a player you can book the court and you can also get access to your video and then we can do AI-powered analysis on it to offer statistics and technique analysis and tips and highlights,” she said.

Their goal is to make it easier to access the personalized analysis that was previously limited to elite players.

The startup closed a $1.8 million seed round in August. It was led by early-stage investor Skyfall Ventures. Backers also include Norwegian pension fund MP Pensjon, former pro tennis player Dekel Valtzer and chess grandmaster Carlsen.

Founded in 2023, SportAI plans to use the funds to further develop its technology, as well as to hire employees and to expand its client base.

Other Services & Software