A leadership team that decides AI will take over part of the work rarely talks about the same condition. One director means that the technology is ready. Another means that the people who must provide oversight are actually authorized to reject something. That difference usually only becomes visible once the project stalls, and by then it is expensive to rewind.
In parts of the company, something is already changing. A task that was entirely human work last year is partly done by AI this year, with a human approving or rejecting it and giving a reason for that. Elsewhere in the same organization, the same kind of task remains entirely human work, not because it could not be done otherwise, but because no one has established who holds the authority to reject. That difference does not lie in the technology. It lies in who has been designated, who can escalate, and who is allowed to hold back the outcome.
Companies where this already works have usually not arranged that by accident. They have established somewhere which role may make which decision when AI delivers an outcome. Companies where it does not yet work often have formulated an ambition, but have not designated anyone responsible for the moment things go wrong.
The ambition test does not start with the technology but with the language. A leadership team says "we want AI to take over the first line of customer contact" and means four different things at once by that: a longer-term vision, an intermediate state that counts as acceptable, a concrete target for the coming year, and the state in which that target must be achieved. As long as those four layers have not been named separately, the leadership team talks past each other without anyone noticing. The difference between an ambition and a goal is often underestimated, and what is the difference between a goal and an ambition shows why that confusion is not merely semantic.
The test measures readiness across eight dimensions, from technical infrastructure to decision rights, and along five confidence gates that follow one another. That does not produce a score you put on a dashboard and then forget. An outcome of the test says something about where the organization stands now, not about where it will stand in a year, and certainly not about whether the ambition is achievable independent of the people who must carry it out. If one of the eight dimensions is insufficiently mapped out — for example because no one knows how mature the data infrastructure is — then the outcome on that point is an assumption, not a measurement. That is marked separately in the test, and it is the reason why why readiness is not a number but a series of questions has its own page: a single number suggests a precision that the underlying data often does not have.
Moreover, the method cannot say what should happen to people whose work changes. That is a decision for the employer, with its own statutory requirements that bind an employer. The test delivers facts about which work can likely be shifted and under which conditions; what an organization does with that regarding personnel falls outside what this method assesses.
An ambition that is achievable on paper stalls for three reasons, and none of the three has anything to do with the AI technology itself.
First, the initiative remains with the department that manages the technology, while the decision rights needed to actually shift the work lie with a different department. How to recognize and bridge that gap is worked out in how do you prevent an AI project from staying stuck with the IT department.
Second, the ambition is technically and operationally achievable, but the compliance requirements were not factored into the planning. A target that already works on the floor can still come to a standstill because a regulator or internal audit function has not yet approved a framework. What to do about that before the project runs aground there is set out in what do you do when the ambition is achievable but compliance is not.
Third, departmental goals work against each other without anyone linking that to the ambition. One department is measured on turnaround time, another on error margin, and an AI application that serves the one undermines the other. How to make that conflict visible before it delays the rollout is set out in what do you do when departmental goals work against each other.
Part of the outcome of the test is sometimes that an ambition is better off waiting a year, not because the technology is lacking but because one of the eight dimensions — often decision rights or data quality — must first grow to match it. That is not a rejection of the ambition, it is a timeline that fits what the organization can currently substantiate. Which signals point to that is set out in which ambition is better postponed for a year.
The outcome of the ambition test is a picture of readiness at a given moment, built from answers that are only as good as the knowledge of the people who give them. If that knowledge is missing on one of the eight dimensions, then it is also missing from the outcome. The test does not tell you what will happen; it tells you where the assumptions lie that must first be tested before a target may reasonably be called a target.
The question of which work in this specific company can actually be shifted to AI, with human oversight or without, is answered per task by the work scan of FTE TO AI.
The free readiness check consists of eight short questions, one per dimension, and delivers a picture of where the organization is furthest along and where it is least far along. The full ambition test, with the four layers, the five confidence gates and the translation back to roles and decision rights, 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.