The use of technology in sports is the subject of passionate debates between traditionalists that believe that technology harms the true nature of sport and technophiles that argue that technology increases transparency. These debates are getting less intense since the development of reliable artificial intelligence (AI) technologies. Tennis provides an illustrative example of how artificial intelligence can be adopted without hurting a sport’s identity: the substitution of human line calling systems (line judges) for electronic ones. In this piece, I make a parallel between this AI driven transition and how AI is changing work in organizations.

Besides the players, three other types of actors are present in a tennis court: chair umpire, ball kids, and line judges. Umpires are responsible for managing the ball kids and line judges and for assuring that players follow the rules. These tasks require interpersonal skills and flexibility to apply the rules when something unexpected happens. Ball kids have a very specific task, picking the balls from the court and delivering them to the players. Unlike the umpire, they do not need special skills or knowledge, but they need to be flexible since it is difficult to predict where the ball ends up after each point. Line judges have a specific set of skills that allows them to keep high levels of concentration and self-discipline under stressful situations. They have a very specialized and repetitive task in which a mistake can be very consequential. The characteristics of these three functions explain why the introduction of electronic calling systems did not affect them equally: line judges were substituted; ball kids were unaffected and for umpires the effect is not straightforward.

The new technology influences how umpires conduct their tasks and how they are trained. Umpires are responsible for other team members, ball kids and line judges, and intervene when necessary (e.g. wrong call from a line judge). Such interventions are sensitive since they question the performance of another team member and line judges’ future calls (often players get angry at this line judge which puts additional pressure on them). With an electronic line calling system, umpires do not need to oversee line judges and can focus on other aspects, such as monitoring the behavior of players.

Traditionally, umpires develop their skills by climbing the task tennis ladder: ball kid, line judge and, finally, umpire. That is, the process to become an umpire relies on experience gained in other tennis tasks. In a world in which line judges are substituted by electronic line calling systems, this training will be disrupted. As more and more tournaments adopt electronic line calling, to be umpires will not be able to gain the necessary experience needed to become high-performance umpires.

The impact of technology on tennis is a good illustration of how technology affects work in organizations. To better understand the increasing rates of task automation, economics and management scholars have relied on a perspective coined by Autor and co-authors as “task approach”. This approach conceptualizes the workplace as a collection of interrelated activities that are typically grouped into three main categories: abstract, routine and manual. Abstract tasks are mainly non-routine cognitive activities like data analysis, creative thinking, and complex communication that require analytical and interpersonal skills. These tasks, despite not being (yet) at risk of being fully automated can be complemented by AI technologies. Despite their non-cognitive nature, manual tasks also cannot be easily taken by AI technologies due to their non-routine nature that requires flexibility. In contrast, routine tasks, defined by their repetitiveness and codifiability, are susceptible to substitution by AI-aided machines.

The characteristics of these three types of tasks present in the workplace mirror the roles present in a tennis court. Umpires, analogous to jobs involving abstract tasks, must have deep knowledge and flexibility, skills that are complementary to technology. Ball kids perform manual tasks that are less susceptible to automation due to their variability and physical nature. Line judges, performing routine tasks, are increasingly threatened by technology, as evidenced by the adoption of the electronic calling systems.

Interestingly, the parallel between tennis and organizations, can also be extended to the indirect effects on managerial jobs. First, automation of routine tasks allows managers to focus their attention on tasks in which humans are (for now) irreplaceable, like personal interactions. This is, of course, dependent on how reliable the technology is perceived. In a context in which it is mature and trustworthy, like the electronic calling system, technology complements and facilitates the managerial task, in a context in which the technology is less reliable, it might lead to poor decision making and additional managerial costs. Second, like in the traditional umpire training system, managers also develop their skills by performing other jobs in organizations. Automation might lead to leaner organizations in which there are less learning opportunities for managers.

The impact of technology in tennis has interesting parallels to how AI is changing work in organizations. In this piece, I claim that the substitution of line judges for electronic line calling systems and the limited direct effect that this technology has on the tasks performed by umpires and ball kids illustrates broader economic trends in job automation linked to AI: professions that rely on routine tasks are more at risk of substitution than professions associated to abstract, and manual tasks. Moreover, I suggest that the transition observed in tennis also shows how AI can indirectly affect professions that at a first glance are not obviously threatened by automation. AI complements abstract activities and allows individuals to focus on activities that depend on their analytical and interpersonal skills. At the same time, excessive automation of routine tasks might limit learning by doing opportunities.

Author: Pedro de Faria - p.m.m.de.faria@rug.nl

Complementary readings from the author:

Broekhuizen, T., Dekker, H., de Faria, P., Firk, S., Nguyen, D. K., & Sofka, W. (2023). AI for managing open innovation: Opportunities, challenges, and a research agenda. Journal of Business Research, 167, 114196.

Fonseca, T., de Faria, P., & Lima, F. (2019). Human capital and innovation: the importance of the optimal organizational task structure. Research Policy, 48(3), 616-627.