If an ambition assumes that work shifts to AI, work also shifts to the people who remain: the oversight, the exceptions, the assessment of what an AI system delivers. The gate "can your people handle it" does not test whether your employees are motivated or loyal. It tests something narrower: does the organization have the skill to assess AI output, reject it with a reason, and adjust it without needing an external party for that. That is a different question than "are they doing their current work well", and that difference is often overlooked.
The three categories of AI-work cut across this. Where AI fully takes over the task, the work of the person involved changes into something else or the task disappears from that role. Where oversight with reason is needed, a new core task arises: not executing, but assessing. And where the work remains human work, little changes — except that the organization must know that this is so, and why.
In some organizations, this distinction has already been made: per task it has been established whether AI takes it over, partly takes it over with oversight, or does not. In other organizations, that distinction exists only in the ambition of management, not in the daily practice of the teams. The difference rarely lies in the technology. It lies in whether someone has taken the time to look per task at what exactly happens, who currently looks at it, and what changes when part of that work shifts.
Companies where this already works have generally also established who may decide that AI output is rejected, and on what grounds. Without that established, a gray zone arises: AI delivers something, no one feels ownership of the assessment, and the quality of the oversight depends on who happens to walk by. What you establish about decision rights in a change determines whether that zone continues to exist or is resolved before the work shifts.
This gate does not ask for an intention or a training plan. It asks for evidence that people are already today doing what the ambition expects of them, on a small scale, demonstrably. Sufficient evidence is, for example: a team that already assesses AI output on a subtask and documents rejecting it with reason. Insufficient evidence is an assumption that people will be able to do this once the time comes, or a training that is planned but not attended.
The difference between those two is measurable in time and in capacity: how many hours have already been freed up by AI takeover, and has that freed-up time been used for oversight of something else, or has that time silently disappeared without anyone actually having redesigned the work. If no one can answer that question, there is no evidence, only an ambition.
An unwelcome outcome is usually: the evidence is missing, or the evidence shows that the oversight is already too thin for the tasks that have already been taken over today. That is a finding, not a conclusion about who should stay or leave. If the outcome of this gate touches on personnel decisions, its own legal requirements apply to that; this test does not provide substantiation for that.
What the test does provide is a reason to revisit the ambition itself before the organization begins it: perhaps the four layers of the ambition — vision, vision state, target, target state — have been set too quickly relative to what is already in place in terms of oversight and assessment. This also touches on other readiness questions: how do you test whether your suppliers can keep up if the oversight partly lies outside the organization, and how do you weigh growth against margin in a strategy if the freed-up capacity is deployed elsewhere than planned.
Under "can your people handle it" lies a more precise question: which work in this company is per task really transferable to AI, and which part continues to require oversight. That question is answered per task with the work scan of FTE TO AI, apart from what the ambition assumes.
The full ambition test — four layers, eight dimensions, five confidence gates, with a translation back to roles and decision rights — is under construction. What is available today: a free readiness check of eight short questions, one per dimension, which in a few minutes gives an impression of where your organization is furthest along and least far along. That impression is a starting point, not a judgment about people, and a logical step up to the question of whether the people behind a task are already doing what the ambition will soon require of them.
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.