← All work signals
Labour signal

The Machine's First Look

AI is entering talent evaluation across football clubs, MLS and US front offices. One Saudi club rules players out on data benchmarks before human judgement applies.

What is it?

Models now make the first cut on players: Al Kholood’s benchmarks rule candidates out, ai.io’s app scores phone drills, and Sportico judges NFL teams near certain to be testing AI in evaluation. How far a score can stand in for a scout’s eye is open.

Why now?

ai.io’s app scores phone clips, and Teamworks says its models cover about 100,000 players across more than 200 leagues, with human judgement applied afterwards at Al Kholood. Vendors now supply that first look to clubs and MLS, while Sportico judges NFL teams almost certainly at least testing AI tools.

Who’s building it?

Teamworks, ai.io and Microsoft, with club and front-office staff deciding how far to rely on them.

Teamworksai.ioMicrosoftClub recruitersFront offices
The full read
Why now, who is building it, where the market goes, and what stays unsettled.
Get weekly →
01 · The market
Who’s in it
Builders — Teamworks, ai.io and Microsoft supply models and tools.
Buyers — football clubs, MLS and NFL front offices testing AI in evaluation.
Gatekeepers — UEFA’s scout programme and club staff who judge players who pass.
What’s at stake
Who gets looked at: a benchmark or app score decides whom clubs invite or rule out.
Local and foreign players face different thresholds at Al Kholood.
Scouts’ role once a model has made the first cut.
What’s unsettled
How a club divides work between a benchmark and a scout’s judgement.
Whether NFL use goes beyond the coding help Walder expects.
What Teamworks’ 70 to 80 per cent transfer success rate does and does not establish.
02 · How the demand matureda progression, not a timeline
Novelty
Clubs and teams try models and an app for scoring players.
Retention
Benchmarks become part of recruitment, set per role at Al Kholood.
Willingness to pay
Teamworks claims 70 to 80 per cent transfer success, against under half for most.
Category arc
A first screen of benchmarks sits before the scout’s judgement at Al Kholood.
03 · Where the market goesthe forks, not a prediction
→Screen then judge: models make the first cut and people judge those who pass, as at Al Kholood and in UEFA’s teaching.
→Phone footage scores decide whom clubs invite, reaching athletes conventional scouting networks overlook.
→Use stays largely in coding assistance, which Seth Walder expects to be by far the biggest NFL use.
→Transfer success and trial counts remain Teamworks’ and ai.io’s own accounts, per the reports.
Everything turns on how much of a player’s worth a score can capture before a person looks, and who checks the vendors’ figures.
04 · The Post-Labor Workforce deskadjacent markets, moving now
The Machine Moderator
AI now asks the follow-up questions.
Salesforce agrees to buy an AI interviewing company.
Research Live
The Predicted Injury
Footballers’ injury forecasts get their first evidence review.
Most reviewed injury-prediction studies rate at high risk of bias.
Journal of International Medical Research
Synthetic Sports
Scanned World Cup players become avatars for offside calls.
FIFA to scan all 1,248 World Cup players into AI avatars
TNT Sports
05 · The Wire
01
A Saudi club drops players who miss data benchmarks.
The National ↗
02
An app turns phone footage into scouting scores.
AGBI ↗
03
Sportico: NFL teams are almost certainly testing AI.
Sportico ↗
The panel on this signal
They agreed on
–Models now sit at the first step of player evaluation, from club benchmarks to a phone app.
–At Al Kholood, people judge players who clear the screen.
–Some figures come from vendors and are reported as their own claims.
The labour economist disagreed on
–At Al Kholood human judgement starts with the model’s shortlist, so a cut-off decides who gets human judgement of potential.
Know more about the method →
✦How this signal was built
01 · Panel
A Delphi panel of 4 AI analysts, each reasoning independently.
→
02 · Debate
They challenge and revise each other’s reads over two rounds.
→
03 · Evidence
Every claim is tied to a cited, linked source on The Wire.
→
04 · Synthesis
A moderator writes it into one read, disagreement and all.