AI has also been used by football clubs like Liverpool FC to analyze their opponents’ teams and playing styles, as well as their strengths and weaknesses. Liverpool FC has said that it is collaborating with the AI arm of one of the world’s largest technology companies to “combine computer vision, statistical learning and game theory” to help spot patterns in the data they collect.

Professional athletes, as well as amateurs, are using “wearable” tracking devices like smart bands and smart rings to track vital health statistics that are then analyzed using intelligent algorithms. Smart rings are used by a number of NBA teams like the Miami Heat and the Detroit Pistons, and athletes like LeBron James, Rory McIlroy and Michael Phelps have all used smart bands, which have even been approved as wearables by various sports leagues including the PGA Tour and Major League Baseball. The PGA Tour used smart bands to provide early indications of whether golfers had contracted COVID-19.

AI is also being used to monitor player movements and identify patterns that may lead to injuries. This information can be used to adjust training routines and prevent injuries. AI imaging is also being used to diagnose patients, including athletes, at a much earlier stage, allowing for pre-emptive care and decreasing the likelihood that risk factors are missed. Software specialists and radiologists have been teaming up for years to come up with algorithmic formulas, teaching computers to identify abnormal parameters that sports teams are not using in the field of injury prevention. In the United States, for the U.S. Food and Drug Administration (FDA) to approve an algorithm involving imaging, it must be accurate 80% to 90% of the time. So far, the FDA has approved about 420 of these for various diseases. For more information in respect of privacy considerations, please see the Data protection and privacy section.

One of the challenges in respect of processing data in real-time (such as real-time blood pressure and individual player GPS data) is the availability of fast and accurate processing, as well as the low-latency movement of data. Faster networking is a crucial aspect of such processing, and as 5G continues to become more ubiquitous, this data processing and movement will become increasingly available and the associated AI solutions will become more commonplace. Formula 1 (F1) is a shining example in respect of real-time data processing and real-time data feedback. Data has completely transformed F1. An F1 car is covered in a vast array of sensors, allowing for 2TB of data to be taken from a car in a single weekend of racing, while a vast amount of this data is passed back to the F1 team in real time. With 5G capabilities and increased real-time processing capabilities, it should become possible to substitute “sportsperson” for “car” to see the potential that real-time AI-powered data analysis has for the future of sports.

Matches: Before and during

AI algorithms can analyze vast amounts of data on teams and players to make accurate predictions about the outcome of a match. At the 2022 FIFA World Cup in Qatar, an international broadcaster’s AI robot correctly predicted the outcome of seven out of eight knockout games. Humans have been using technology in an attempt to predict the outcome of sports matches for some time; however, a new breed of AI-driven algorithms is proving to be more successful than the old methods. Using predictive algorithms and computer vision models, AI is unlocking a new level of accuracy in respect of outcome prediction in sports.

At the 2022 World Cup, FIFA implemented a complex AI-powered offside detection system. A dozen strategically placed, AI-connected cameras continuously collected data from 29 specific points on players’ bodies, while a sensor, which sits at the center of the ball, relayed the ball’s exact location to the officials at a rate of approximately 500 times per second. The system was designed to alert a control room if it caught a player offside, and the humans in the control room would then relay the information back to the on-field officials.

Similar to the offside technology mentioned above, cricket and tennis have a history of using the Hawk-Eye detection system (which has been around since 2001). In 2020, in response to the need to reduce the number of people on court (because of COVID-19), the U.S. Open tennis tournament replaced human line judges on 15 of 17 match courts with Hawk-Eye Live, an advanced system that made automated line calls in real time. The Hawk-Eye Live system featured 18 cameras, six of which were used by a review official to monitor foot faults. The system uses recorded voices to make its calls, which shout “out,” “fault” or “foot fault.” While Hawk-Eye does not use “neural networks” and is not considered a “deep learning” AI system, as the tennis ball is generally moving too quickly to “see” whether the ball is in or out, it is generally considered an AI system as it uses a statistical model, based on inputs provided by the cameras, to decide – on the basis of probability – where the ball was when it touched the court. Newer versions of Hawk-Eye include real-time AI analysis, which is advanced enough to automatically make a call on sporting decisions as complex as football’s offside rule. This technology is already being used by analysts in the NBA and Italy’s Serie A football league. Football fans (and long-suffering assistant referees) hope to see these new technologies remove uncertainty from offside decisions going forward.

Broadcasting/fan experiences

Modern sports media strategies go beyond television, encompassing official applications, social media platforms and direct-to-consumer (DTC) platforms. By using new digital platforms and new AI-powered technologies, it is possible to enhance fan engagement in existing markets and attract new fans. In 2022, the sports technology industry witnessed deals worth $90 billion, with fan engagement constituting 40% of overall activity and media and broadcasting representing the two most rapidly expanding segments. Sports are now competing with many forms of entertainment for viewers, and the industry is engaged in a battle to deliver the right content to the right user on the right platform at the right time.

AI technologies have the potential to enhance traditional sports broadcasting production methods, addressing the challenges of latency and content delivery speed. Companies are working with global sports organizations and rights holders to create automatic, personalized videos in real time. AI technology can analyze various elements of a match, such as crowd noise, scoreboards and commentator voices to select the most important video segments and seamlessly compile them into a bespoke highlights package. Additionally, if a federation is restricted to showing only one minute of on-field action, AI can optimize that 60-second window by including as many key moments as possible, helping to maximize engagement opportunities.

Visual recognition technology can also tailor packages for sponsors, featuring moments that coincide with their logo being displayed on shirts, advertising hoardings or anywhere on the screen, enhancing commercial partnership opportunities.

AI is being used to enhance the fan experience in sports, providing real-time statistics and analysis (as well as highlights, as mentioned above) during games. This can include analysis of associated social media activity to identify trending topics and players, as well as providing personalized content to fans based on their interests and preferences. The NBA has collaborated with a technology partner to create an AI-powered chatbot to engage with fans both during and outside of the relevant game, providing fans with real-time updates on games, player statistics and other relevant information. Chatbots may also be used outside of office hours to respond to fan inquiries in relation to commonly asked questions about broadcast times, ticket prices and ticket availability.

Computer vision models may be used to detect specific areas on a sports field, recognize objects and monitor player movements. Additionally, AI-supported optical character recognition (OCR) and audio detection are being deployed alongside cognitive technologies like facial recognition and spatial awareness to increase the information a fan is able to engage with during a match, contributing to increased fan engagement. These systems rely on the deployment of integrated cameras, drones and devices positioned at various locations on the field, sometimes providing a comprehensive 360-degree view of the game and the ability to zoom in on key moments.

Sports broadcasters and their technology providers can leverage extensive object and AI libraries from incumbent technology companies to identify and manipulate objects within video frames. This can be used to enhance virtual advertising and fan engagement and improve the quality of creative graphics.