Sport’s AI Needs an Owner. Not Another Demo.

The demo is the easy part. A screen, a convincing output, a room briefly imagining how much faster everything could move. The harder question arrives afterwards: whose budget pays for this, whose working day changes, and who answers when the output is wrong? For sport, that is the more useful innovation brief. Not whether AI looks impressive in isolation, but whether an organisation can make it useful without leaving responsibility somewhere between the performance department, the commercial team and the software supplier.

Follow the budget

The State of AI in Sports 2026, presented by Teamworks on Front Office Sports, offers a revealing snapshot. Among the organisations surveyed, 24% had a specific AI budget. Another 38% paid through existing software or operating budgets. And 29% did not know how their AI spending was funded. The remaining 9% were not spending on AI yet.

These figures describe the surveyed organisations, not every club or sports business. The supplied excerpt does not establish the sample or methodology, and its commercial context matters. Still, the budget split raises a practical question worth asking inside any organisation: can the people adopting a tool explain what they are committing to?

Paying through an existing software budget is not automatically a problem. A separate AI budget is not automatically a strategy. The distinction that matters is visibility. Someone should be able to explain the total commitment, the intended benefit and the conditions under which the experiment stops.

That includes more than the subscription. A sensible assessment should count staff time spent checking outputs, preparing inputs and adapting the workflow. Otherwise, a tool can look inexpensive while the work around it remains invisible. Before asking whether AI saves money, establish which work it is supposed to change.

Give someone the whole job

The report also notes that AI use is often siloed by department. Asked who makes organisational decisions about AI, one of the joint-most-common answers was that each department decides for itself.

Local experimentation has a logic. The people doing a job are well placed to identify its frustrations. A commercial team should not need a grand transformation programme to test a better way of preparing an internal document. But local freedom needs a route into shared decisions when tools, information or responsibilities cross departmental boundaries.

The answer is not necessarily a new executive title. It is a named owner with enough authority to connect the people involved. That person should know who approves the spend, who checks the output, who can access the inputs and who decides whether the test continues.

An AI pilot is not a strategy until someone owns what happens after the demo.

Ownership should not mean pretending one person understands every professional discipline. Coaches, commercial staff, event producers and other specialists still need to judge whether an output is useful in their work. The owner’s job is to make those judgements part of the process, rather than an inconvenient final hurdle.

Test a task, not a promise

How should a sports organisation start using AI? Choose a bounded task, record how it works today, and agree on what improvement would justify continuing. Test the workflow, not the presentation.

An internal event briefing, for example, could be a proposed starting point: limited in scope, reviewable by its producers and separate from decisions about an athlete’s health or selection. This is a suggested test, not evidence of a particular organisation’s deployment. Its value would depend on whether it produces a better working document after human review, not merely a faster first draft.

Decide the evaluation criteria before seeing the results. Look at accuracy, review time, usefulness to the next person in the chain and the effort required to correct mistakes. If drafting gets faster but checking becomes slower, that trade-off belongs in the assessment.

Also define the handover. A successful test should leave behind instructions, a responsible person and a clear review point. A failed test should leave a useful explanation. Neither outcome needs a launch campaign. The aim is to improve a job, not to manufacture a technology story.

What it means

At DUĒLL we see the strongest commercial question here as one of professional practice: does the technology help people deliver something better for the people sport serves?

For brands, an AI partnership brief should specify the useful outcome and the person accountable for delivering it. Do not confuse access to a tool with a reason to sponsor it. For clubs, protect room for departmental experiments while making spending and evaluation visible across the organisation.

For athletes and their representatives, ask who is responsible whenever a proposed workflow touches their information or professional decisions. For communities, insist that success includes the quality of the experience, not just the organiser’s administrative savings.

The business takeaway is straightforward: buy against a defined job, test against an agreed standard, and give someone responsibility for the result. The most credible innovation announcement may be the one made only after the work has improved.