FIFA’s Semi-Automated Offside Technology has processed every offside ruling since Qatar 2022 using the same generic graphical body shape, applied identically to a 5ft 7in winger and a 6ft 4in centre-forward. At World Cup 2026, that geometric shortcut ends: 1,248 players across 48 squads will each be scanned in approximately one second, producing personalised 3D anatomical models embedded directly into the VAR decision pipeline.
By David Findlay, Founder of KiqIQ.
How 1,248 One-Second Body Scans Are Rebuilding FIFA Offside Technology for World Cup 2026
The SAOT Geometry Problem
FIFA’s Semi-Automated Offside Technology, first deployed at the 2022 World Cup in Qatar, tracks player positions using optical cameras and a body-part skeletal model. The system calculates whether any part of an attacker’s body that can legally score is beyond the last defender at the moment the ball is played.
The core limitation: SAOT currently applies a standardised graphical representation to every player on the pitch. The existing system uses generic body shapes regardless of a player’s actual physical proportions. A 5ft 8in striker and a 6ft 5in goalkeeper are processed through the same geometric template. At fine margins, the tolerances that separate a legal goal from a disallowed one, individual body dimensions matter in ways a generic model cannot account for.
The Scanning Solution
FIFA and technology partner Lenovo have built a one-second scanning process to resolve this. Each player enters a specialised chamber during pre-tournament photo sessions and undergoes a rapid full-body scan. The process captures what FIFA describes as highly accurate body-part dimensions, generating a personalized 3D model unique to that player’s physical profile.
With 48 teams each carrying 26-man squads, the total scope is 1,248 individual scans. FIFA’s technical teams will embed these models within the SAOT system so that each offside calculation reflects the actual skeletal and muscular geometry of the specific player being assessed, not a generic approximation.
VAR Integration and Host Broadcast
The 3D models serve a dual function. Beyond supporting VAR decision-making, the personalized avatars will appear in the host broadcast. FIFA confirmed the models will allow offside rulings to be displayed more realistically and in a more engaging way to fans at stadiums and to viewers around the world.
In the current SAOT broadcast visualization, a stylised stick-figure outline illustrates the offside call. The World Cup 2026 system replaces that with a rendering that reflects each player’s actual body shape: arm length, shoulder width, and torso proportions specific to the individual involved in the decision.
Football AI Pro and the Wider Intelligence Layer
The 3D avatar upgrade operates alongside Football AI Pro, built on FIFA’s proprietary Football Language Model. The system analyzes over 2,000 different metrics and processes millions of data points per game, giving coaching staff, analysts, and players access to validated tactical insights in text, video, graph, and 3D visualisation formats.
Football AI Pro is designed to support all 48 participating teams, enabling pre-match opposition analysis, in-session question-and-answer with a generative AI assistant, and post-match breakdown with integrated video clips. The Football Language Model was trained on hundreds of millions of FIFA-owned data points across competition history.
Intercontinental Cup Test
The avatar-based SAOT system was trialled at the FIFA Intercontinental Cup. Players from CR Flamengo and Pyramids FC were scanned ahead of their match, and the system ran throughout the game, validating the scanning workflow and the model’s integration with match officiating systems. That trial is the primary evidence base for World Cup deployment confidence.
What Changes at Fine Margins
The practical effect of personalized geometry is most relevant at borderline decisions. SAOT already reduced the time taken to reach an offside decision compared to the manual process. The body scan upgrade addresses a different problem: it narrows the margin of geometric error in the model itself, not the speed of the decision.
For a system where goals have been disallowed by centimetres, the difference between a generic shoulder width and a player’s actual shoulder width is a design-level accuracy improvement. FIFA’s stated goal is to make decisions faster and ensure a clear understanding by everyone of the ruling, including players and coaches watching the broadcast visualization in real time.
Frequently Asked Questions
What is Semi-Automated Offside Technology?
SAOT is FIFA’s computer vision system for detecting offside positions. It uses optical cameras and body-part tracking to calculate whether an attacker’s limb that can legally score was beyond the last defender when the ball was played. It was first deployed at the 2022 FIFA World Cup in Qatar.
Why does body shape affect offside accuracy?
A player’s physical dimensions affect the precise position of limbs relative to the defensive line. Generic models apply the same body geometry to all players. At fine margins, actual arm length, shoulder width, and torso depth determine the correct call. Personalized models remove the geometric approximation from the calculation.
How long does each player scan take?
FIFA has confirmed each scan takes approximately one second, conducted inside a specialised chamber during pre-tournament media sessions. Scans will cover all 1,248 players before the tournament begins.
How many players will be scanned?
All 1,248 players from the 26-man squads of 48 participating nations will receive individual 3D scans. The system was tested at the FIFA Intercontinental Cup with players from CR Flamengo and Pyramids FC.
What is the Football Language Model?
FIFA’s Football Language Model is a proprietary domain-specific AI model trained on hundreds of millions of FIFA-owned data points. It powers Football AI Pro, the generative AI assistant available to all 48 teams at World Cup 2026, analyzing over 2,000 metrics per match.
Sources
- Irish Times: World Cup players to have lifelike AI avatars for VAR offside decisions, January 7, 2026
- TechNode: CES 2026 AI-powered 3D digital avatars to support refereeing at FIFA World Cup 2026, January 7, 2026
- Inside World Football: FIFA set to use AI technology to improve semi-automatic offside experience, January 8, 2026
- Artificial Intelligence News: FIFA World Cup 2026 will be the most AI-driven tournament ever, March 12, 2026
- Sports Video Group: FIFA President Infantino on AI impact at 2026 FIFA World Cup, January 12, 2026