If you watched the 2026 FIFA World Cup, you most likely noticed the now-familiar ritual: a aim, a celebration, then a referee touching an earpiece whereas followers and gamers look ahead to the decision from the video assistant referee, or VAR.
That ritual shouldn’t be distinctive to soccer. Baseball has a replay center in New York, and different sports have additionally moved once-human calls into review systems.
The hope is easy: Use higher expertise to yield higher calls.
Soccer authorities anticipated extra cameras, replay angles and monitoring information to mitigate referee error, restrict subjectivity and strengthen perceptions of equity. But VAR has created new arguments amongst gamers, coaches, followers and commentators: Should VAR have intervened? Was the identical normal for when to intervene utilized persistently? Did the method itself really feel honest?
Watching the World Cup as a decision-making researcher, I couldn’t assist however see VAR as greater than a refereeing device. It regarded like a case research in technology-assisted decision-making, with parallels to the growing use of AI in organizational decision-making.
VAR produced a set of mismatches between what the expertise promised and how folks skilled it. It left fans disappointed and created new disputes about judgment, course of and belief. These tensions supply a helpful lens for understanding AI adoption extra broadly.
AP Photo/Martin Meissner
Better measurement shouldn’t be higher judgment
Some choices are measurement issues: Was the player offside? Did the ball cross the road? Did contact happen? Technology is great at these questions as a result of cameras, sensors and information cut back errors that come from folks not seeing clearly.
But many choices are judgment issues: Was the contact sufficient for a penalty? Was the deal with reckless? Was the referee’s authentic name clearly improper? Those questions contain interpretation, context and requirements. A digital camera can present that contact occurred. It can’t determine how that contact ought to be judged.
The similar problem seems when organizations use AI to help choices. AI can course of information, predict patterns and classify circumstances. But a physician, supervisor, choose or trainer nonetheless must determine what the output means, how a lot weight to provide it and what values are at stake. Research on human-AI collaboration makes an identical level: AI is usually strongest when paired with human judgment quite than handled as a full substitute for it.
Subjectivity doesn’t disappear – it strikes
Before VAR, arguments centered on what the referee noticed. After VAR, arguments typically concentrate on how the method works. When does VAR intervene? How far again can officers review the play? What counts as “clear and obvious” video proof?
Subjectivity remains to be there. It has moved from the referee’s eyes to the foundations, thresholds and governance of the review system. Baseball reveals an identical sample: Replay didn’t take away judgment from the sport. It changed which performs might be reviewed, how challenges labored and the place judgment entered the method.
AP Photo/John Minchillo
AI programs create the identical shift. Organizations typically think about AI as a technique to take away human discretion. In follow, discretion reappears in new locations: deciding which mannequin to make use of, what information to coach on, what error price is appropriate and who’s accountable when the system fails.
More precision can cut back belief
People typically assume that better precision creates better belief. Sometimes it does. But it may possibly elevate expectations sooner than it reduces ambiguity. If expertise can measure an offside choice by centimeters, followers could count on each choice to really feel equally sure. When penalty calls or pink playing cards stay debatable, folks can turn out to be pissed off.
Research on why people avoid using algorithms finds one thing related: People can lose confidence in algorithms after seeing them make errors, even after they carry out nicely general. Accuracy alone doesn’t assure legitimacy.
That discovering issues for AI choice programs. An organization could use AI to make hiring or efficiency analysis extra goal. But if candidates or staff see the system as inconsistent, biased or unfair, belief can fall as an alternative of rise. When organizations body AI as a silver bullet, they elevate expectations the expertise can’t all the time meet. The consequence could be deeper frustration and sooner lack of belief.
Visionhaus/Getty Images
Where expertise ends and judgment begins
Management research means that AI adoption as we speak shouldn’t be a easy alternative between people and machines. The higher query is whether or not a choice ought to be automated, augmented or left to human judgment.
Measurement issues are the strongest candidates for automation: AI can scan paperwork, detect patterns, classify circumstances or flag anomalies faster and more consistently than people can. Interpretation issues name for human-AI collaboration: AI can present info, choices or suggestions, however folks nonetheless must suppose about the context of the choice, how a lot uncertainty is concerned and the implications the choice could have. And some choices stay deeply human. Questions involving equity, accountability, values or that means can’t be handed over to expertise with out altering the character of the choice itself – from one in every of human judgment to algorithmic evaluation.
VAR makes this boundary seen in sport. AI is now forcing organizations to confront the identical boundary throughout high-stakes choices. The key query in AI adoption shouldn’t be merely, “Can AI make this decision?” It is, “Which parts of the decision should AI make, and which parts should remain a matter of human judgment?”