--- title: "AI literacy in sport: two populations in clubs, three in federations" url: https://zerofive.ai/en/blog/compliance/ai-literacy-sports-clubs-training canonical: https://zerofive.ai/en/blog/compliance/ai-literacy-sports-clubs-training language: en published: 2026-07-10 updated: 2026-09-23 author: "ZeroFive.AI" tags: AI literacy, sport, clubs, match analyst, Annex III abstract: "Technical and commercial staff fall under different regimes. The match analyst as critical role and the problem of supplier-owned data." --- # AI literacy in sport: two populations in clubs, three in federations **In short.** A sports club has two populations using AI for opposite purposes: technical staff, working on data from contracted athletes and therefore on systems classified as high risk, and commercial staff, working on content and the public. Training has to be separated, because the obligations differ. The critical role is the match analyst, who in most clubs has no formalised position on the org chart and almost never appears in a training plan. In federations and sports promotion bodies the populations become three, because the administrative one joins in, the largest of the three and the most exposed on the handling of member data. In sports organisations, mandatory training exists for safety, anti-doping and sometimes sporting integrity, and doesn't exist as a structured framework. That is the main difference from banks and insurers, and it changes the starting point: here it isn't about extending a system, it's about building a minimal one. ## Two populations, two regimes Technical staff work on systems touching contracted athletes. Scouting, performance evaluation, workload management: when those systems affect contractual decisions they fall among the worker management systems of Annex III, because a professional athlete is a worker. Commercial staff work on content, ticketing and the relationship with the public, and on an asset clubs treat as marketing material rather than personal data: the supporter database. Segmenting season-ticket holders, personalising an offer, choosing who receives a campaign are processing activities under the GDPR, with legal bases to be stated and an additional question about minors, who make up a significant share of a sporting audience. On the AI Act side the obligations concern transparency, mainly Article 50, and the risk level is low. Treating both populations with the same module produces a plan that serves neither. Whoever analyses performance needs to know when a model is outside its range; whoever runs the club's channels needs to know what must be declared as artificially generated. ## In federations the populations become three In a federation or a sports promotion body a third population joins the other two, and it is the largest: the people working on registrations, affiliations, courses and administrative procedures. These are processes touching tens of thousands of people, largely minors, and they include medical fitness certificates, which are health data. Two areas deserve specific attention. The first is the training of coaches and match officials: systems that determine admission to a course, evaluate an exam or monitor a test fall within the education and vocational training area of Annex III, where the classification is high risk. Several federations have digitised those courses without realising they had entered that perimeter. The second concerns the role of the body itself. Where a federation acts as a public-law body, high-risk systems carry further obligations, starting with the fundamental rights impact assessment. It is a legal check worth doing in advance, because it changes how much documentation is required. ## The match analyst and their absence from training plans This is the figure at the centre of the question and the one almost no club has formalised. They receive data from external analytics suppliers, interpret it, and their readings feed decisions on line-ups, playing time and transfers. For this role the competence evidence has to be individual, because this is the person effectively exercising human oversight over the system. They need to recognise when a supplier's data is unreliable, how the model behaves on small samples, and when technical judgement should override the number. The other two roles needing individual evidence are the head of performance, accountable for the use of health data, and whoever in management approves the adoption of a platform. ## When the data doesn't belong to the club In clubs almost every system comes from suppliers: scouting platforms, wearables, tracking systems, video analysis tools. The club is a deployer and rarely has visibility on how the model is built or when it gets updated. This has a direct consequence for training: you cannot teach people how a system works when its workings are unknown. What can and must be done is to train them on what the system claims to do, what its known limits are, and what to do when the output doesn't add up. Requesting documentation from the supplier therefore becomes a training requirement as well as a contractual one. ## What to build and on what timeline | Audience | Content | Frequency | |---|---|---| | Technical staff and match analysts | Systems in use, known limits, when the data doesn't hold, how a departure is documented | On joining and at every platform change | | Performance and medical staff | Handling of health data, legal basis, who accesses what | Annual | | Commercial and communications staff | Disclosure obligations for generated content, limits on image use | Annual | | Marketing, CRM and ticketing | Legal basis for processing supporter data, profiling, consent, handling of minors | Annual | | Registrations, affiliations and administration | Member data, minors, fitness certificates, what an automated system may decide | Annual | | Training and qualifications | Systems assessing admissions and exams for coaches and match officials, Annex III obligations | On joining and at every platform change | | Management | What gets approved when signing off a platform adoption | On joining and before each new adoption | | Stadium operators | What is permitted on access control systems | Before each season | The last row is the one clubs underestimate most. Recognition systems in venues are the area with the strongest constraints, and whoever operates them needs to know what is and isn't allowed. ## When to train, inside a season Training in a club can't follow a corporate calendar, because the season sets the pace. Two windows work: pre-season training camp, when staff are together and matches haven't started, and the winter break. Outside those windows structured training competes with the fixture list and loses. It's a practical constraint worth accepting during planning rather than discovering in October. ## Where to start The prerequisite is the system inventory, which in a club mostly means mapping suppliers and understanding which data leaves the organisation. The guide is in [the AI system inventory](/en/blog/compliance/ai-system-inventory-iso-42001). From there the two populations get separated and the matrix built, described in [a roles-competence matrix for AI](/en/blog/compliance/roles-competence-matrix-ai). The full requirements are in [clause 7 explained without jargon](/en/blog/compliance/clause-7-iso-42001-explained). The sector's regulatory picture is on the [AI governance for clubs and sports organisations](/en/sectors/sport) page. Our approach to role-based tracks is on the [AI Training](/en/services/ai-training) page. To review your club's situation, you can [book a meeting](https://calendly.com/fabiolalli/zerofive).