■ SAFE RISK ■ Sports & Entertainment
No, because the product is human limits being tested in public. A machine that runs faster is a machine, not a competitor — the entire commercial logic of sport depends on the athlete being someone you could theoretically be.
“Nobody pays to watch robots play. Unless it's BattleBots. That's different.”
Our AI replacement risk score — how we score jobs
The visible part of a professional athlete's job — competing — is a small slice of the week. The rest is periodised training blocks, video review, physio and rehab, nutrition compliance, sleep and load monitoring, media obligations, sponsor activations, and the endless management of a body that is being deliberately pushed toward its failure point. Contract value increasingly tracks audience attention as much as performance, which is why social presence is now part of the job description rather than an accessory to it.
AI has already taken over the analytical layer. Computer-vision tracking systems chart every player's positioning and off-ball movement; load-management models flag injury risk from GPS and heart-rate variability data; opponent-scouting reports that took analysts a week are generated in minutes. Recruitment and draft models rank prospects. In some sports, automated officiating has replaced human line judges outright. Synthetic media is starting to encroach on the commercial side — a sponsor can generate an athlete's likeness for an ad without booking a shoot, which affects earnings even though it does not affect the sport.
None of that threatens the athlete, because nobody has ever paid to watch optimal play. They pay for tribal allegiance, uncertainty and the spectacle of a person doing something extraordinary at cost to themselves. Chess engines have crushed grandmasters for decades and human chess is more popular than ever. The genuine risks are elsewhere: shorter careers as data-driven squads rotate more aggressively, harsher performance surveillance, and a narrowing window for athletes whose numbers dip. Our risk score of 22 reflects that the job changes shape, gets more monitored and more media-dependent, but does not vanish.
Automatability: our editorial assessment of current and near-term AI capability
Disruption is already here but it lands on the support staff, not the athletes. Analysts, scouts and line officials have been displaced through the 2020s. For athletes themselves, expect tighter data-driven selection, more automated officiating and synthetic likeness deals reshaping endorsement income by 2030. The core career remains viable well past 2040; the historical threats to it — injury, age, and roster maths — are unchanged.
They can exist and some do, but they compete with motorsport and esports, not with human athletics. Sport sells identity and jeopardy — the sense that a person like you might fail publicly. A machine breaking a record is an engineering result, not a story. Chess proved the pattern: engines dominate, and human competition grew anyway.
Mostly through measurement. Tracking and load models decide who plays, who rests and who gets renewed, and they flag decline earlier than a coach's eye would. That can extend careers through better injury prevention or end them faster when the numbers turn. Athletes now negotiate against analytics as much as against scouts.
Some of it. Brands can already generate synthetic presenters and, with rights, reuse an athlete's likeness across markets without a new shoot. That reduces appearance fees while increasing licensing value for the biggest names. Mid-tier athletes lose the most, which makes likeness clauses one of the more important things in a modern contract.
A career that is more surveilled, more media-driven and possibly shorter on the field but longer as a brand. Learn to read your own performance data, protect your likeness rights, and build a following that is yours rather than the club's. The athletes with the most leverage in 2035 will be the ones who own their audience.