An AI ambition usually starts with an assumption about the customer: faster response, lower price, a service that does not yet exist. That assumption is rarely tested separately. It travels along with the ambition and is taken as proven along the way. This gate asks one question, separate from the rest: is the customer demand that the ambition assumes the customer demand that actually exists, and is that demand formulated sharply enough that capacity can be set against it.
That is different from the question of whether AI can technically take over the task. A task can be taken over and still not match what the customer is asking for. Conversely, a customer demand can be entirely correct while the work behind it is still mostly human work, with AI at most a preparatory step under supervision. The gate tests the demand itself, not the execution.
The shift is not in the customer demand but in what can be set against that demand. As long as work was entirely human work, capacity was a fixed quantity: so many employees, so many hours, so much turnaround time. Now that part of that work can be taken over by AI -- fully, partly with review that approves or rejects with reason, or still not at all -- what a customer demand means for the organisation answering it also shifts.
A demand that used to require a team of ten now requires, at some companies, a team of three plus oversight of a system that handles most of the standard cases. At other companies, with the same customer demand and the same sector, nothing changes, because the data is not in order, the process is not standardised enough, or the exceptions are too numerous to leave to oversight. The difference is rarely in ambition. It lies in what is already in place: how the work is broken down, how consistently it is carried out, and how well it can be recognised which case is an exception and which is not.
Anyone who tests the customer demand without taking this layer into account is testing an assumption about the market without including the assumption about their own organisation. Both must hold true before the ambition is feasible at all.
Sufficient evidence is not market research with a percentage attached. It is three things that together demonstrate that the customer demand is sharp enough to set capacity against:
If one of these three is missing, the customer demand has not been tested but assumed. That is not necessarily a problem -- many ambitions start with an assumption -- but it is something different from a gate that establishes readiness.
The outcome may be that the customer demand is less well substantiated than the ambition assumes, or that the underlying work is less transferable than thought. Both outcomes are information, not a verdict. The ambition can remain in place; the path towards it then changes shape. Sometimes that means a longer lead time, sometimes a different order of dimensions, sometimes a reformulation of the vision state so that it matches what can be demonstrated today.
Where the outcome touches on who will still be doing the work, separate statutory requirements apply that fall outside this test. The ambition test establishes what the organisation can do, not what an employer must do with that outcome.
This customer demand test does not stand apart from the rest. It is connected to the question of whether the work is margin-viable if AI takes it over, to the operational feasibility of the ambition in day-to-day execution, and to the legal side of work that is assessed or carried out by a system. A customer demand that stands strong on its own can still get stuck on one of those other dimensions.
The question of which work in this company can genuinely be taken over by AI, task by task, is answered with the work scan from FTE TO AI.
Anyone who wants to know whether their own vision is sharp enough to pass this test can first look at how a vague vision is made concrete in a vision state and target. Anyone wondering who will decide on the outcome of this test can find a starting point at what needs to be recorded about decision rights when work changes.
The free readiness check gives a first picture: eight short questions, one per dimension, with an outcome that shows where the organisation is furthest along and where it is least so. The full ambition test, with the four layers and the five confidence gates, is under construction.
Vertel wat u wilt bereiken, dan kijken we samen wat daarvoor moet staan.
Answers come from this site’s knowledge base. Not tailored advice, and not a scan of your company.